Loading…

Sign up or log in to bookmark your favorites and sync them to your phone or calendar.

Tuesday, May 12
 

8:15am

8:29am

ALL TIMES LISTED IN PACIFIC TIME ZONE
All times listed on OCP website and in SCHED are in the Pacific time zone. When you log into Intrado for the live show, times will be shown in your own time zone.

Tuesday May 12, 2020 8:29am - 8:30am

8:30am

OCP Experience Center
OCP Accelerator Module, OAI ( Facebook, Hyve Design Solutions, Inspur, Xilinx)
OAI will demonstrate currently available OAMs as well as show functional implementations in multiple OAI systems.

OpenEDGE (Artesyn Embedded Power, ASUS, BelPower Solutions & Protection, Flex, Inventus Power, MiTAC, NOKIA)
OCP Open Edge project creates open specifications, standards and reference designs for Edge Computing and Telco use cases that require small form factor server designs that enable high performance and low latency computing. Participants of the project are across the ecosystem including Telco Service Providers, OCP solution providers, ODM/OEM manufacturers, and key technology providers. Nokia has contributed the initial Open Edge solution that creates the basis for interoperable designs.

RunBMC (Dropbox, Intel, Vertiv)
This is a demo of hardware modules following OCP RunBMC specification. The first video delivered by authors of the specification from Dropbox features the demo of HS2500 Prototype Module and HS2500 Production Module. This video also presents the standalone RunBMC Bring Up Vehicle and RunBMC-compatible Olympus chassis with Nuvoton Poleg RunBMC Module. The second video delivered by Intel presents FPGA-based RunBMC module featuring Intel® Cyclone V and Intel® MAX10. The video demonstrates the integration of this module in standard Intel 2S Skylake-X platform by using RunBMC Bring Up Vehicle card from Dropbox. In this video you can also see OpenBMC boot on Intel RunBMC module and basic management of the platform with WebUI, SOL and Remote KVM.

Open System Firmware, OSF (Facebook, Wiwynn, HP Enterprise, Intel and ITRenew).
Open system firmware is an open development project, the goal of which is to allow OCP owners to 'own their firmware' -- to move the point of control of firmware to the system owner. Starting in March, 2021, OCP badging for servers will require that systems support OSF. We are exploring multiple ways to accomplish this with components including Hostboot, UEFI, coreboot, and LinuxBoot.

OpenRMC (Inspur)
The Demo will show the establishment of OpenRMC Northbound 1.0 profile execution on top of the bundle of OpenRMC FW & HW (RMC) contributed to OCP as a complete reference solution to address the fulfilment of user's requirements/demand for a comprehensive management of an entire rack system and beyond via RESTful API through remote management portal/client console with detailed WebUI and paving the way for exploring future OpenRMC v2.0.

SEBA Solution (MITAC and EdgeCore)
SEBA solution is one of ONF project and with Edgecore and MiTAC, you can have the OCP hardware ready for SEBA.

OCP NIC 3.0 group (Amphenol, Broadcom, Dell Technologies, Facebook, Intel, Marvell Semiconductor, NVIDIA Networking - Mellanox, Quanta, TE Connectivity, UNH-IOL, Wiwynn Corporation)
The goal of this subgroup is to form a healthy OCP NIC 3.0 ecosystem through a supportive, informative and inclusive community.

96Tb DDC based Disaggregated Router ( Credo, DriveNets, Edgecore Networks, Keysight Technologies, UfiSpace )
The Open Compute Project (OCP) has selected UfiSpace, Credo and DriveNets to give the first ever live demonstration of a Distributed, Disaggregated Chassis (DDC) model as proposed to the OCP by ATT. This demonstration is the largest live demonstration of a distributed router cluster ever given in public. The technology is aligned with the vision of TIP (Telecom Infra Project) who also invited DriveNets and UfiSpace to advise on their Disaggregated Open Router Initiative and ensure its alignment with the OCP's efforts around network disaggregation.

Tuesday May 12, 2020 8:30am - 4:00pm
Experience Center

8:30am

Expo Hall Open
Tuesday May 12, 2020 8:30am - 4:00pm
Expo Hall

9:00am

Welcome to the Virtual Summit
As we pivot from our annual in-person Summit to a virtual event, we highlight what to expect over the next four days, and share the unique opportunities within the virtual environment.

Speakers
avatar for Dirk Van Slyke

Dirk Van Slyke

Director of Marketing, Director, Marketing & Events
Dirk began his career over thirty years ago working in brand management leadership for full-service, creatively-driven branding agencies such as The Richards Group and GSD&M. He has led successful marketing initiatives in a variety of industries, including ecommerce, healthcare, fashion... Read More →


Tuesday May 12, 2020 9:00am - 9:02am
Keynote Theater

9:02am

Message from the OCP Foundation Board
OCP Chairman of the Board, Mark Roenigk, will review the global impact of the OCP Community over the past year, highlight some of the benefits of the new virtual experience, introduce new board members, Kushagra Vaid and Jim Hawkins, new co-chairs, Jessica Gullbrand and Jeff Catlin, plus a surprise announcement!

Speakers
avatar for Mark Roenigk

Mark Roenigk

OCP Chairman of the Board, Facebook
As head of hardware engineering, Mark Roenigk is responsible for the building, scaling and innovation of the infrastructure necessary to support Facebook’s mission to give people the power to build community and bring the world closer together. Facebook builds efficient and flexible... Read More →


Tuesday May 12, 2020 9:02am - 9:10am
Keynote Theater

9:10am

OCP Future Technologies Symposium Winner Annoucement
The Symposium program chairs will announce the winning abstract and award the $10,000 prize!

Speakers
avatar for Lesya Dymyd

Lesya Dymyd

Innovation Engineer, 2CRSI
Primary lead for the successful execution of an event, including communication with academic institutions and leading the Program Committee. Lesya holds a doctorate degree from the University of Strasbourg and was a visiting researcher at ETH Zurich. She is currently a strategic innovation... Read More →
avatar for Allan Smith

Allan Smith

Area 404 Lab Manager, Facebook
Allan is a graduate of the US Air Force Academy and ESC Toulouse. He is currently the Lab Manager for Area 404, Facebook’s hardware prototyping laboratory. Prior to joining Facebook, Allan led Rapid Response Space Works efforts with NASA Ames Research Center.


Tuesday May 12, 2020 9:10am - 9:13am
Keynote Theater

9:13am

OCP Highlights, Partnerships, and a Look Ahead
CEO Rocky Bullock will share the direction of the organization, latest market forecast, and announce partnerships that were recently formalized.

Speakers
avatar for Rocky Bullock

Rocky Bullock

CEO, OCP
Rocky has more than 40 years of experience in the technology industry with expertise in professional services, systems integration, product design, new product introductions, technology development, sales, operations, manufacturing and logistics. He has held executive and senior manager... Read More →


Tuesday May 12, 2020 9:13am - 9:22am
Keynote Theater

9:22am

OCP: Past, Present and Future - Presented by Facebook
As we mark the transition to new decade, we reflect on the growth of OCP over past decade. Starting with a bold idea of doing hardware designs in an open ecosystem, OCP has come a long way. This talk will reflect on some of the early motivations of how the OCP came about, where we are today touching on the theme of “Open for All” and chart the potential course for the next 10 years ahead for the industry. The talk will also cover some of the challenges we face from Facebook perspective which perhaps has common pain points for all of us in the infrastructure space and hope to inspire next decade of designs, innovations and contributions into OCP that lead to building all data centers with “OCP First” philosophy.  

Speakers
avatar for Katharine Schmidtke

Katharine Schmidtke

Director of Sourcing for ASICS and Custom Silicon, Facebook
Dr. Katharine Schmidtke is Director of Sourcing for ASICS and Custom Silicon at Facebook. Over the past five years she led Facebook’s Optical Technology strategy and worked closely with OCP to specify the 100G-CWDM4-OCP optical transceiver optimized for data center applications... Read More →


Tuesday May 12, 2020 9:22am - 9:42am
Keynote Theater

9:42am

A Global Circular IT Industry Opportunity and Imperative
The global IT industry is responsible for 4% of global emissions and on track to double by 2025. The OCP community is doing a great job driving advances in operational energy efficiency. But this is only part of the puzzle, one third of the puzzle in fact. To truly assess and address the environmental impact of data center equipment, it is imperative to include both pre-use (embodied) and post-EOL costs of the massive scaling of our collective infrastructure. In this session we will explore the true total cost of equipment, and the opportunity represented by circular data centers to redefine lifetimes, and maximize both financial value and sustainability. Operationalizing circular data centers means enabling a global, circular IT hardware industry, and to do so catalyzes both financial and environmental opportunity, and democratized access to growth. We’ll talk about what this entails, what’s happening now, and what is at stake.

Speakers
avatar for Ali Fenn

Ali Fenn

President, ITRenew
Ali Fenn is the President at ITRenew, where she oversees all revenue and leads the Company’s circular data center initiatives, including market development and business model innovation. Open hardware platforms and open source software innovation are the critical foundation underpinning... Read More →


Tuesday May 12, 2020 9:42am - 10:00am
Keynote Theater

10:00am

Adopt Compute, AI, Storage, and Total Solutions - Presented by Inspur
OCP adoption has grown tremendously in the past couple of years.  IHS Markit reported an impressive $2.56B revenue, excluding revenue from OCP board member companies, and expected to surpass $10B by 2022.   With the explosion of cloud application usage and the fast-growing servers for AI workloads, reducing power consumption, deploying energy-efficiency solutions, and achieving the promise of AI require computing platforms that deliver high-performance, robustness and scalability.  It becomes increasingly important for datacenters to embrace openness to enable interoperability and to respond more quickly and cost-effectively to market demands.  Alan will share Inspur use cases utilizing open technologies for both high-performance computing and AI solutions.  He will explain value propositions to customers and collaborations with key partners.  Alan will also share Inspur’s latest collaborations in new areas for storage and networking.  Together, we embrace openness and accelerate adoption of open technologies.  

Speakers
avatar for Alan Chang

Alan Chang

Sr. Product Director, BU Cloud Computing, Deputy GM, IEI Server Product Line Dept., Inspur
With over 10+ years’ experience in various roles in server product planning & account management in OxM. Having worked as everything from software programmer fresh out of university to hardware product planning in major system supplier, Alan has developed a diverse skillset that... Read More →


Tuesday May 12, 2020 10:00am - 10:20am
Keynote Theater

10:20am

A New Multi-Industry Collaboration for an Open Storage Platform - Presented by Samsung
Today’s age of hyper-convergence, hyper-scale cloud service providers, and hybrid cloud-edge computing opens vast possibilities for service owners. But it also creates a problem: the need to quickly and efficiently process enormous amounts of data in systems that are robust, reliable, and easily implemented on any scale. Open Storage Platforms (OSPs) have great potential as a fundamental element of these systems, but a truly effective OSP must reflect the needs not only of service owners but also hyper-scale public cloud infrastructure and diverse end users.
 
For this reason, Samsung has launched a multi-industry collaborative program with a system manufacturer and data center to develop, test, deploy, and contribute to an open-source OSP. Samsung’s long-standing expertise in flash memory and advanced NVMe technology is a perfect complement to the knowledge and perspective of our partners. The open-source approach is well-suited to this type of project, where all involved parties can benefit from faster innovation, broader standardization, and better economic return as cloud-scale deployments of the infrastructure will be possible. Instead of deploying full server racks in single increments, these types of systems can be implemented by the tens to hundreds of thousands.
 
In this keynote session, Samsung will share the vision for key aspects of this open-source OSP and introduce the development partners who will together undertake this community-based problem-solving effort.

Speakers
avatar for Jongyoul Lee

Jongyoul Lee

Software Development Team Leader, Memory Division Semiconductor Business, Samsung Electronics Co.,Ltd.
Jongyoul Lee  Software Development Team Leader  Memory Division Semiconductor Business  Samsung electronics Co., Ltd.Jongyoul Lee is Senior Vice President of Samsung Memory Business and currently leads S/W development team in Solution division.Jongyoul has 17 years’ experience... Read More →


Tuesday May 12, 2020 10:20am - 10:40am
Keynote Theater

10:40am

Accelerating Innovation with New Solutions from Cloud to Edge - Presented by Intel
Being customer focused requires us to understand more deeply infrastructure challenges and problems.  Our approach is shifting from providing components to delivering value across the full platform lifecycle.  Intel Vice President, Jennifer Huffstetler will discuss how Intel’s latest OCP contributions are enabling optimized solutions for growing workloads from edge to cloud.

Speakers
avatar for Jennifer Huffstetler

Jennifer Huffstetler

Vice President, Data Platforms Group & General Manager, Strategy, Intel
Jennifer M. Huffstetler is a vice president of the Data Platforms Group (DPG) and General Manager of Strategy in the Xeon and Memory Group. In this role she is responsible for driving the overall Xeon PnL, building Data Platform technology & business strategies to drive future growth... Read More →


Tuesday May 12, 2020 10:40am - 10:51am
Keynote Theater

10:51am

Emerging Networking Scenarios with SONiC - Presented by Microsoft
SONiC (Software for Open Networking in the Cloud) has been innovating fast and going beyond public clouds. In this keynote, we will present a key breakthrough in SONiC, contributions from ever-growing community, why customers choose SONiC and what enables you to adopt SONiC.  

Speakers
avatar for Dave Maltz

Dave Maltz

Distinguished Engineer, Microsoft
Dr. David A. Maltz leads Azure's Physical Network team, which is responsible for developing, deploying, and operating the software and network devices that connect the servers of Microsoft's largest services, including the Azure Public Cloud and Bing.  We write the code for the software... Read More →


Tuesday May 12, 2020 10:51am - 11:13am
Keynote Theater

11:20am

Panel Discussion: Open Networking and SONiC
In this live panel, we will have industry leaders from various high tech industries to share their vision on open networking. We will have an open discussion on where SONiC is as a platform, what advice they could give to companies looking to OSS solutions, and what is their blueprint of infrastructure as going into the future.

Moderators
avatar for Dave Maltz

Dave Maltz

Distinguished Engineer, Microsoft
Dr. David A. Maltz leads Azure's Physical Network team, which is responsible for developing, deploying, and operating the software and network devices that connect the servers of Microsoft's largest services, including the Azure Public Cloud and Bing.  We write the code for the software... Read More →

Speakers
QC

Quaid Campbell

Director, Cloud Platform and Infrastructure, T-Mobile
avatar for Yiu Lee

Yiu Lee

Vice President of Architecture, Comcast
Yiu Lee is Vice President of Architecture at Comcast. His team is responsible for Comcast IP network architecture.
avatar for Thomas Soupault

Thomas Soupault

Head of Networking, Criteo
Thomas is the Network Manager for Criteo, responsible of the production network for Criteo business. He oversees the area of network load-balancing, edge, and data center.
avatar for Parantap Lahiri

Parantap Lahiri

VP NW and DC Engineering, eBay
Parantap currently leads network, datacenter and server hardware engineering at eBay as a vice president. In past, Parantap led network engineering and architecture at Microsoft for several years before his years at Juniper Networks where he led solution engineering for Contrail SDN... Read More →
avatar for Pablo Espinosa

Pablo Espinosa

VP of Engineering, Target
Infrastructure Technologist with 20+ years of combined experience in eCommerce, Enterprise and CDN environments across a variety of business verticals. Organizational leader with a proven track record of building and retaining high performance organizations with a focus on designing... Read More →


Tuesday May 12, 2020 11:20am - 12:00pm
Keynote Theater

11:59am

[ON DEMAND] Advancing Open Networking with Juniper
Juniper continues its commitment to increasing network openness, flexibility and reliability. This session covers how open source development, such as SONiC on Juniper platforms and integrating additional functionality into SONiC, aligns with Juniper’s strategy and direction. Join Juniper’s Principal Product Manager, Sirisha Paturi as she discusses Juniper’s commitment to driving the SONiC ecosystem and providing the best solution at all layers of a disaggregated stack including:
• The industry’s first integration of MPLS and MACSec with SONiC
• The advantages of Juniper’s SONiC implementation in a modular chassis
• Additional integration point benefits with Juniper’s disaggregated software such as a containerized control plane

Speakers
avatar for Sirisha Paturi

Sirisha Paturi

Principal Product Manager, Juniper Networks
Sirisha has over 18 years of experience in the networking industry. She has been with Juniper for over 9 years and has driven product strategy and roadmap for success in changing market conditions.She currently manages the PTX modular portfolio.


Tuesday May 12, 2020 11:59am - 11:59am
Expo Hall Stage

11:59am

[ON DEMAND] Air-assisted Liquid Cooled Olympus rack/server using A-L RDHX - presented by Microsoft
Direct Contact Liquid Cooling provides a solution to higher chip and rack density needs. However, many of today’s data centers are built based on aircooled designs such as free cooled-adiabatic architecture. As an intermediate step for higher chip power enablement, OCP ACS group suggested the use of A-L heat exchangers. This enables higher rack density, overall reduced power use and peak performance potential without the need to retrofit facility. This session will discuss the use of rear door heat exchangers as a cold plate enabler for the Olympus server and rack.

Speakers
avatar for Brandon Rubenstein

Brandon Rubenstein

Principal Platform Engineer, Microsoft
I've had the privilege to contribute to Microsoft's success as an engineering director, a design architect and a lead designer. Most recently I've managed the mechanical and power teams responsible for defining, designing, and maintaining Microsoft's cloud business hardware and datacenter... Read More →
avatar for Husam Alissa

Husam Alissa

Sr. Engineer, Microsoft
Husam is a technical lead working for Microsoft Data Center Advanced Development Team with focus on next generation technologies for hardware and infrastructure.
avatar for Robert Lankston

Robert Lankston

Sr. HW Engineer, Microsoft
Robert is a Sr. Mechanical Engineer in Microsoft's Cloud Hardware Engineering team with a focus on thermal design of servers. At Microsoft, Robert contributed to the thermal design and architecture of the Olympus infrastructure as well as many of the server platforms that integrate... Read More →


Tuesday May 12, 2020 11:59am - 11:59am
Expo Hall Stage

11:59am

[ON DEMAND] Battery Backup Unit: More Integrated or Modular Building Block - presented by LITE-ON
Battery Backup Unit (BBU) is becoming more popular than traditional UPS approach, both in enterprise servers and in large scale data centers.  LITEON has offered various BBU designs, some are integrated, some with modular approach.  This presentation will discus design considerations, trade-offs, as well as directions in future applications.

Speakers
avatar for Sam Ye

Sam Ye

Associate VP, Technical Sales and Marketing, LITEON
Sam Ye got his Ph.D. from Virginia Tech in 2000. He worked at GE Global Research Center (2000~ 2005), and DELL (2005~2006).  he joined LITEON Technology Corp. in 2006 as Director of Engineering, and recently served as Associate VP of Technical Sales and Marketing. His background... Read More →


Tuesday May 12, 2020 11:59am - 11:59am
Expo Hall Stage

11:59am

[ON DEMAND] Can Block Storage Devices Converge on NVMe? - presented by Seagate
During the last ~40 years, some of us have seen many evolutions of block storage protocols come and go: IDE/ATA, SCSI, PATA, P-SCSI, SATA, USB, FCP, SAS … NVMe. Each protocol optimized specific Enterprise, Consumer or Cloud storage device solution or feature set. Regardless of the various storage media types and characteristics, is it finally possible for the storage industry to unify and consolidate around NVMe as the optimal block storage protocol? … Why? … When?

Speakers
avatar for Mohamad H El-Batal

Mohamad H El-Batal

Office of the CTO, Seagate Technology
Mohamad El-Batal is the Seagate Enterprise Data Solutions (EDS) Chief Technologist, and part of the overall Seagate Office of the CTO team. He is given the opportunity to shape the Seagate EDS strategy and future storage product technology roadmap.In his career Mohamad lead a team... Read More →


Tuesday May 12, 2020 11:59am - 11:59am
Expo Hall Stage

11:59am

[ON DEMAND] Closed Loop Cooling for High Performance Compute Systems - presented by Hyve
With the ongoing increase in power dissipation of processors for high performance computing applications (artificial intelligence, deep learning, machine learning, etc.), the need for more efficient and practical cooling systems is clear. This presentation demonstrates the use of a scalable hybrid liquid cooling system which utilizes cold plates and cooling fluid in direct contact with high heat generating components to transfer heat away from the components to the liquid-to-air heat exchangers. This solution is applicable to large scale data centers, allowing for higher power density, efficiency, and performance without having the need to update facility cooling infrastructure.

Speakers
avatar for Alekhya Addagatla

Alekhya Addagatla

Thermal Engineer, Hyve Solutions
Alekhya Addagatla is a Thermal Engineer at Hyve Solutions, working on high-density servers solutions and next generation hardware technologies for data centers. She has over 4 years’ experience building hardware systems for compute, storage and networking systems for Hyperscale/Cloud... Read More →


Tuesday May 12, 2020 11:59am - 11:59am
Expo Hall Stage

11:59am

[ON DEMAND] Data Centers Today, Trends, and Challenges- presented by QCT
Data Center Trend
- Speed, Performance and Power Trends
- AI & Edge
Data Center Challenge
- Commodity TDP increases (CPU, Memeory, NVMe SSD, NIC)
- System level Air cooling limitations
Future Data Center Architecture
- Edge servers to support 5G workloads
- HW and Resource utilization
- High Density system design
- Liquid Cooling (NCHC, PUE, etc)

Speakers
avatar for Dustin Wu

Dustin Wu

Director, Telco Solution Engineering, QCT
Dustin is currently working with QCT (Quanta Cloud Technology) as the Director of Telco Solution Engineering, primarily responsible for overseeing the resolution of technical sales issues for Service Provider customers. On this role, he also manages the QCT’s Telco Ecosystem Lab... Read More →


Tuesday May 12, 2020 11:59am - 11:59am
Expo Hall Stage

11:59am

[ON DEMAND] Enhancing SONiC for Enterprise Data Centers - presented Dell Technologies
Since its introduction in 2016, SONiC has gained support by multiple large cloud providers. However, with its growth there remains several challenges impeding a broader reach of implementations in the areas of ease of deployment, operation and support. This session will outline several recent activities where Dell Technologies is working with the SONiC community to address these areas.

Speakers
avatar for Prathyaya Bhandarkar

Prathyaya Bhandarkar

Global Architect, Dell Technologies
Prathyaya Bhandarkar is a Global Architect in with Dell Technologies’s Networking Product Group. Prathyaya brings 10+ years of experience in Networking, Software Engineering, and Product Management. Prathyaya has broad experience across the Technology Industry in Computer Networking... Read More →


Tuesday May 12, 2020 11:59am - 11:59am
Expo Hall Stage

11:59am

[ON DEMAND] FB Data Center Networking View: Deployment and Beyond - presented by Facebook
Facebook has been driving change within data center networking space over last several years via its active contributions in OCP Networking workgroup. This talk shares our next stage of our journey in deployment and operationalization of large scale networks and challenges there in. The talk will survey few critical areas involving operations, migrations, efficiency and achieving performance at scale.

Speakers
avatar for Omar Baldonado

Omar Baldonado

Director of Software Engineering, Infra Network, Facebook
Omar Baldonado is the co-lead of the OCP networking project. At Facebook, he is part of the networking team, responsible for large-scale software systems that model, automate, monitor, and control Facebook’s network. The network domains include Facebook’s data centers; global... Read More →


Tuesday May 12, 2020 11:59am - 11:59am
Expo Hall Stage

11:59am

[ON DEMAND] Finding the Goldilocks Zone in Data Center Networking - presented by Xilinx
Software Defined Networking has transformed Data Center Networking in several profound ways.  First, it has broken down traditional boundaries of where functionality exists across network elements.  Second, it has vastly increased the functional richness and diversity of features available to data center operators.  And third, it has made possible for the first time the development and deployment of new networking features at the speed of software.

But this transformation has also brought about massive technical challenges and unanswered questions.  What is the most optimal way to partition networking functionality across switches, NICs, and Server CPUs?  When and where does it make sense to apply acceleration techniques?  What types of acceleration technologies provide the best performance while maintaining the flexibility and development agility required by SDN?  

In case you haven’t noticed, many players in the networking industry have set out to tackle some of these difficult problems. This is creating a renaissance in domain specific networking architectures. In this session we will present Xilinx’s vision regarding this important new trend and explain how FPGA technology will play an important and critical role.

Speakers
SS

Scott Schweitzer

Accelerator Marketing, Xilinx


Tuesday May 12, 2020 11:59am - 11:59am
Expo Hall Stage

11:59am

[ON DEMAND] Going SONiC - Together - presented by NVIDIA Networking – Mellanox
With SONiC gaining traction in data centers around the world, it has never been a better time to adopt this fully open source network operating system. Every new SONiC release adds critical functionality. But don’t go SONiC alone. Join other cloud innovators who have leveraged NVIDIA Networking – Mellanox's full ASIC to Protocol (A2P) Support for SONiC.

Speakers
avatar for David Iles

David Iles

Sr. Director, Ethernet Switch, Mellanox
David Iles, Senior Director of Ethernet Switching at Mellanox Technologies:Over the last 20 years, David has held leadership roles at 3COM, Cisco Systems, Nortel Networks, and IBM where he pioneered advanced network technologies including High Speed Ethernet, Layer 4-7 switching... Read More →


Tuesday May 12, 2020 11:59am - 11:59am
Expo Hall Stage

11:59am

[ON DEMAND] Improve Scalability and Application Performance with PCIe® 4.0 SSDs - presented by KIOXIA
Next-generation cloud and IoT applications, including machine learning (MI), artificial intelligence (AI) and 4K/8K media, are driving ever-increasing data intensive workloads. Emerging technologies, such as 5G networks, 400Gb Ethernet and NVMe-oF™ software, are accelerating the demand for higher-performance server and storage infrastructures to support these workloads. PCIe 4.0 technology will ease I/O bottlenecks by doubling the PCIe interconnect speed, paving the way for improvements in resource efficiency, node scaling and overall system performance.

KIOXIA America, Inc. (formerly Toshiba Memory America, Inc.) is the US-based subsidiary of KIOXIA Corporation, a leading worldwide supplier of flash memory and solid state drives. KIOXIA is the inventor of flash memory and an industry leader in enterprise, data center and client SSDs. Leading the industry with PCIe 4.0 SSD development, KIOXIA will highlight its data center and enterprise SSD portfolio and explain how PCIe 4.0 technology can address bottlenecks in data-intensive workloads.

Speakers
avatar for Neville Ichhaporia

Neville Ichhaporia

Senior Director, Marketing and Product Management, Data Center and Client Division, SSD BU, KIOXIA
Neville Ichhaporia is senior director for marketing and product management of the Data Center and Client SSD Division at KIOXIA America, Inc. where he leads the team focused on defining product strategy, product planning, and business management for KIOXIA’s product portfolio targeting... Read More →


Tuesday May 12, 2020 11:59am - 11:59am
Expo Hall Stage

11:59am

[ON DEMAND] Improving Application Latency and Predictability Through Application Device Queues (ADQ) - Presented by Intel
The greater demands of workloads on network bandwidth and achieving application and network latency requirements is ever more challenging. Application Device Queues (ADQ) increases application response time predictability, thereby reducing application latency and improving throughput to better meet customer service-level agreements (SLAs).

Speakers
avatar for Todd Koelling

Todd Koelling

Communications Segment Marketing Manager & leader of Application Device Queues (ADQ) ISV enablement, Intel
Todd Koelling is Communications Segment Marketing Manager and leader of Application Device Queues (ADQ) ISV enablement at Intel.  He has 25+ years industry experience with a platform and solutions focus, encompassing technical, marketing and strategic planning roles in processors... Read More →


Tuesday May 12, 2020 11:59am - 11:59am
Expo Hall Stage

11:59am

[ON DEMAND] Innovation in Open Storage Solutions - presented by Inspur
Inspur is going into the technical details, value pro, and use cases of the open storage platform and our collaboration with partners.

Speakers
avatar for Alan Chang

Alan Chang

Sr. Product Director, BU Cloud Computing, Deputy GM, IEI Server Product Line Dept., Inspur
With over 10+ years’ experience in various roles in server product planning & account management in OxM. Having worked as everything from software programmer fresh out of university to hardware product planning in major system supplier, Alan has developed a diverse skillset that... Read More →


Tuesday May 12, 2020 11:59am - 11:59am
Expo Hall Stage

11:59am

[ON DEMAND] Open from Cloud to Edge, Our Solutions for the Seamless Transition - presented by Wiwynn
As of July 2019, every major cellular carriers in the US was offering 5G services, thought not available everywhere yet. The fast deployment of 5G network pushes the Edge computing to work hand in hand with cloud to provide a flexible solution for real time imagine processing, analyzing and data collection.

While Wiwynn has abundant cloud product and technologies expertise, we like to make sure we have the right building block in the pipeline thus we can continue to support our cloud customers when they are ready to ride the 5G tide and adding the edge infrastructure.

We will be showing our products and roadmap for cloud and edge solutions. We are demonstrating the latest cutting-edge OCP platforms, and the edge solutions offering highly integrated, future proof, compact and power efficient power efficiencies.


Speakers
avatar for Steven Hwang

Steven Hwang

Executive Director, Wiwynn
Steven is the Executive Director for Sales Enablement for Wiwynn. He is experienced as a solution architects for cloud infrastructure. Along with the capacity to capture the market vibe and identify growth opportunities, Steven’s other strengths lie in his ability to build relationship... Read More →


Tuesday May 12, 2020 11:59am - 11:59am
Expo Hall Stage

11:59am

[ON DEMAND] Open Uncomplicated - presented by ITRenew
We have the proof that “Open for All “isn’t just a utopian vision, its real and ready today: pre-configured, plug-in-ready rack-scale solutions, built on open architecture. No assembly. No guesswork. No army of engineers required. Now anyone can power their data center like a hyperscaler while dropping their TCO and upping their sustainability game. Come learn how.

Speakers
avatar for Erik Riedel

Erik Riedel

SVP of Compute & Storage Solutions, ITRenew/Sesame
Long-time engineer & engineering leader in infrastructure & storage - interested in bare metal k8s, hardware, and performance at scale.


Tuesday May 12, 2020 11:59am - 11:59am
Expo Hall Stage

11:59am

[ON DEMAND] Rittal’s Experience and Innovations in OCP Solutions
Rittal has been a member of the OCP since 2012 and has played a key role in the development of the open rack. Learn more about the new OCP V3 rack, but also about the new back door cooling (ACS Door HEX Group), which is based on an innovative direct chip cooling. It is now possible to implement and cool high-density HPC nodes in an existing data center.

Speakers
avatar for Resul Altinkilic

Resul Altinkilic

Project Manager Global Key Accounts IT, Rittal
Project Management Global Key Accounts IT, working on High Performance Computing / Cooling and also on Edge Solutions. Involved in the OCP Community since 2017 (Open Compute Project) and the development of OCP Rear Door Solutions.Former Project Manager for the Automotive and automatization... Read More →
avatar for Andy Gill

Andy Gill

VP Engineering & Global IT Key Accounts, Rittal
Most recently became responsible for Global IT Key Accounts within RITTAL whilst maintaining engineering responsibility for rack developments, specialising in hyperscale design and global role out. Involved in the OCP community for 7 years playing a part in developing ORV2 & 3. He... Read More →


Tuesday May 12, 2020 11:59am - 11:59am
Expo Hall Stage

11:59am

[ON DEMAND] Shell Gas-to-Liquid (GTL) Technology for Immersion Cooling - presented by Shell & Asperitas
Presenting how the Shell Immersion cooling fluid is manufactured, key chemical/physical properties, tests and main value proposition and benefits for immersion cooling applications and industry.

Speakers
avatar for Eduardo de Azevedo

Eduardo de Azevedo

Product Application Specialist, Shell
Eduardo de Azevedo is a Product Application Specialist at Shell Global Commercial Technology, based in Hamburg – Germany. He is also leading the development of Shell Immersion Cooling Fluid for Computer Data Centres.


Tuesday May 12, 2020 11:59am - 11:59am
Expo Hall Stage

11:59am

[ON DEMAND] Silicon Motion SSD Controller Focus on Data Center and Storage Trends
E1.S is the form factor winner for 1 U server designs migrating from M.2 based on SSD scalability, flexibility and thermal efficiency.  During our discussion, we will be highlighting our findings with our most recent PCIe Gen NVMe E.1S SSD designs.

Speakers
avatar for Gary Adams

Gary Adams

Senior Director of Marketing, Silicon Motion
Let's talk about making Enterprise SSD Controllers and Solutions better...


Tuesday May 12, 2020 11:59am - 11:59am
Expo Hall Stage

11:59am

[ON DEMAND] Supermicro Solutions: Upping the performance ante while keeping IT Green
In this session, the presenter leads the audience through various challenges facing the IT industry and how Supermicro’s phenomenal innovations for the last 26 years, in compute and storage are catering to the ever increasing data center performance requirements while keeping it all secure, open standards and GREEN.

Speakers
avatar for Vik Malyala

Vik Malyala

SVP, FAE & Business Development, Supermicro
As a Senior Vice President of FAE & Business Development for Supermicro, Vik Malyala manages WW Field Engineering, Solutions & Technology Enablement teams responsible for understanding the large enterprise and service provider customers’ requirements and deploying the optimized... Read More →


Tuesday May 12, 2020 11:59am - 11:59am
Expo Hall Stage

11:59am

[ON DEMAND] Zoned Namespaces: From Archival to I/O Predictability - presented by Samsung
Zones Namespaces (ZNS) represent the first step towards the standardization of Open-Channel SSD concepts in NVMe.  Specifically, ZNS brings the ability to implement data placement policies in the host, thus providing a mechanism to address the log-on-log problem and its implications: high write-amplification factor (WAF), tail-latencies and higher cost.  Following its counter-part in the high capacity HDD world - Shingled Magnetic Recording (SMR) HDDs -, ZNS is initially targeted at archival workloads.  However, we believe that ZNS can also address the other core concept in the Open-Channel SSD architecture: I/O Predictability.  In this short paper, we cover the fundamental concepts on ZNS, the mechanisms that make it attractive for archival purposes, and how it can be extended to provide I/O predictability.

Speakers
avatar for Javier Gonzalez

Javier Gonzalez

Pricipal Software Engineer, Samsung
Javier is a Principal Software Engineer at Samsung Electronics, where he founded and leads SSDR (Samsung Electronics Denmark Research). He is a contributor to the Linux kernel and other open-source projects. His expertise is in building new ecosystems for upcoming technologies. He... Read More →


Tuesday May 12, 2020 11:59am - 11:59am
Expo Hall Stage

12:00pm

Air-Assisted Liquid Cooling: Coldplate Enablement in Aircooled Facilities
Projected trends in component power and density are beginning to push the limits of traditional air cooling. Closed-loop liquid cooling with effective liquid-to-air heat exchange (in short, air-assisted liquid cooling) is a potential approach to extending cooling capability of existing air-cooled facilities. Different rack-level approaches to air-assisted liquid cooling (AALC) will be previewed. The optimized and standardized RPU (reservoir and pumping unit; imagine a CDU without it’s plate heat exchanger) will form the core of any AALC solution and specific focus is placed on defining this component – form-factor, interfaces, features, etc. Short-term goals for this area involve initializing an OCP group to drive specification for and standardization of the RPU.

Speakers
avatar for Cam Turner

Cam Turner

Director, Hyperscale Solutions, CooIIT Systems
CoolIT Systems Inc. is a world leader in direct contact liquid cooling for the Data Center market. CoolIT\'s Rack DCC platform is a modular, rack-based, advanced cooling solution that allows for dramatic increases in rack densities, component performance, and efficiencies. This vendor... Read More →
avatar for Husam Alissa

Husam Alissa

Sr. Engineer, Microsoft
Husam is a technical lead working for Microsoft Data Center Advanced Development Team with focus on next generation technologies for hardware and infrastructure.
avatar for John E. Fernandes

John E. Fernandes

Thermal Engineer, Facebook


Tuesday May 12, 2020 12:00pm - 12:30pm
EW: Advanced Cooling Solutions

12:00pm

Panel Discussion: Open D2D PHYs Panel: One PHY to Rule Them All?
The interest in die-to-die (D2D) interconnect has grown with the increase in interest in
chiplet-based design. Correspondingly, interest has grown in open technologies for
die-to-die communication. 2019 alone saw the introduction of two new die to die
technologies from the OCP, the Bunch of Wires and Open HBI. This follows the release
of the AIB from the Intel and the XSR SerDes from the OIF.

This panel will discuss:
The role for each open PHY technology and the potential overlap between them. How
these PHY technologies can be made widely available.
Identify attributes that need to be/can be common across these PHYs such as management,
test interfaces and a data abstraction.
The role of these open PHY technologies in an OCP marketplace

Moderators
MK

Mark Kuemerle

Fellow, Integrated Systems Architecture, Marvell

Speakers
RC

Rishi Chugh

Vice President, Cadence
avatar for Kenneth Ma

Kenneth Ma

Architect, Xilinx
Kenneth Ma is a Principal Architect at Xilinx. He brings more than 25 years of International standards, Strategic planning, System Architecture and Entrepreneurial experience. Kenneth has held senior technical positions at Nvidia, Broadcom and SK Hynix America. He has previously served... Read More →
avatar for Manuel Mota

Manuel Mota

Sr. Product Marketing Manager, Synopsys
Manuel Mota is the product marketing manager responsible for DesignWare Die to Die Interface IP portfolio. He brings more than 18 years of technical and marketing experience to his position, having held similar responsibilities for the Analog and Wireless IP products. Prior to Synopsys... Read More →
avatar for Ramin Farjad

Ramin Farjad

CTO/VP Networking/Automotive PHYs, Marvell
CT & VP of Networking/PHYs at Marvell Semiconductor, in charge of developing multi-GHz connectivity solutions for autonomous vehicles, hyperscale data centers, and Enterprise networks. Chairman of technical committee at Networking for Autonomous Vehicles (NAV) Alliance. Proposed signaling... Read More →


Tuesday May 12, 2020 12:00pm - 1:00pm
EW: Open Domain-Specific Architecture

12:00pm

Panel Discussion: CSIS Security
In this session, a panel of cloud security experts from the Cloud Security Industry Summit (CSIS, www.cloudsecurityindustrysummit.org) will dicuss top of mind hardware security and supply chain challenges, from a cloud providers point of view, share best practices to address those challenges, and present opportunities for collaboration with the OCP community.

Speakers
avatar for Yigal Edery

Yigal Edery

VP Products, Kameleon
At Microsoft, I held several GPM roles, including cyber security products such as Forefront Threat Management gateway. In my last role, I was the PM for Azure hardware security. Today, I am the VP products of Kameleon, an Israeli startup innovating in the hardware security space... Read More →
avatar for Ben Stoltz

Ben Stoltz

Senior Software Enginerr, Google
Supply chain security.
avatar for Roksana Golizadeh Mojarad

Roksana Golizadeh Mojarad

Senior Cloud Security Architect, Intel
avatar for Matt King

Matt King

Oracle
avatar for Rob Wood

Rob Wood

VP, Hardware and Embedded Security Services, NCC Group
Rob Wood is the VP for the Hardware and Embedded Security Services practice at NCC Group. His career in embedded devices spans two decades, having worked at both BlackBerry and Motorola Mobility in roles focussed on embedded software development, product firmware and hardware security... Read More →
avatar for Tobias Langbein

Tobias Langbein

Enterprise Security Architect, ZKB
At ZKB, a major Swiss bank, Tobias is responsible for the security architecture and roadmap. He is mainly concerned with cyber security, cloud security and resilience and provides strategic guidance and prioritization. Before that he has spent 15 years working on cloud and security... Read More →


Tuesday May 12, 2020 12:00pm - 1:00pm
EW: Security

12:30pm

ACS - Door HEX Step 3
As a part of the ACS Door HEX group, we would like to present the Door HEX Step 2 compatible to the specification. Technical details, feasibility etc.

Speakers
avatar for Resul Altinkilic

Resul Altinkilic

Project Manager Global Key Accounts IT, Rittal
Project Management Global Key Accounts IT, working on High Performance Computing / Cooling and also on Edge Solutions. Involved in the OCP Community since 2017 (Open Compute Project) and the development of OCP Rear Door Solutions.Former Project Manager for the Automotive and automatization... Read More →
avatar for Tim Shedd

Tim Shedd

Director Product Management, ZutaCore
Dr. Timothy A. Shedd is currently Director of Product Management for ZutaCore, Inc. Most recently, he was the Director of the Graduate Program, Supervisor of Entrepreneurship Programs and an Associate Professor of Mechanical Engineering at Florida Polytechnic University. Prior, he... Read More →


Tuesday May 12, 2020 12:30pm - 1:00pm
EW: Advanced Cooling Solutions

1:00pm

Standalone Rack Level Liquid Cooling Solution
The cooling for data center becomes more and more challenging as the silicon power increases significantly. Liquid cooling would be one of the solutions to meet the upcoming high power demand, while maintaining high cooling efficiency. A standalone rack level liquid cooling is one of the liquid cooling solution with high flexibility and requests no data center infrastructure change. In this solution, the heat from high power components is transferred by coolant from cold plate to the rear door heat exchanger, and then the heat is finally rejected by air. The other lower power components are cooled by air.
For hardware, the design of key elements will be standardized and compatible to ORV3 and 19” rack. For software, rack management software will base on Redfish schemas to develop for interoperability and cooling efficiency optimization at run-time by continuously adjusting liquid flow speed, output of rear fan door, and fan speed of individual computer systems. It could be deployed to the existing data centers without any facility change as an interim, efficiency solution before the deployment with facility coolant. In this presentation, Wiwynn and Facebook will co-present the standalone rack level liquid cooling solution.

Speakers
avatar for Yin Hang

Yin Hang

Thermal Engineer, Facebook
AL

Albert Li

Thermal Engineer, Wiwynn


Tuesday May 12, 2020 1:00pm - 1:30pm
EW: Advanced Cooling Solutions

1:00pm

Sinian: A Heterogeneous Computing Framework for OCP-Compliant Accelerators
OAM or ODSA complaint accelerators have much more efficient interconnection between various accelerators. Therefore they should have much larger potential than the traditional architectures when computation tasks are partitioned and run them on different accelerators in parallel. Due to low communication cost, the software optimization strategies and scheduling heuristics also need to be adjusted accordingly. In this talk, we introduce Sinian, an open source heterogeneous framework to enable heterogeneity-aware computation parallelism to maximize the performance for OCP-compliant accelerators. Task partitioning can be based on efficient architecture cost models, or deeply customized by applications.

Speakers
avatar for Weifeng Zhang

Weifeng Zhang

Chief Scientist of Heterogeneous Computing, Alibaba Group USA
Dr. Weifeng Zhang is the Chief Scientist of Heterogeneous Computing at Alibaba Cloud Intelligence Infrastructure, responsible for acceleration and optimization for machine learning and other data-intensive applications. He received a PhD in Computer Science & Engineering from UC San... Read More →


Tuesday May 12, 2020 1:00pm - 1:30pm
EW: Open Domain-Specific Architecture

1:00pm

In-Scope Security Threats
We will outline the threats that are in scope and out of scope for the OCP Security group's specification. 

Speakers
avatar for Yigal Edery

Yigal Edery

VP Products, Kameleon
At Microsoft, I held several GPM roles, including cyber security products such as Forefront Threat Management gateway. In my last role, I was the PM for Azure hardware security. Today, I am the VP products of Kameleon, an Israeli startup innovating in the hardware security space... Read More →


Tuesday May 12, 2020 1:00pm - 1:30pm
EW: Security

1:30pm

Air-Assisted Liquid Cooled Olympus Rack/Server Using A-L RDHX
Direct Contact Liquid Cooling provides a solution to higher chip and rack density needs. However, many of today’s data centers are built based on aircooled designs such as free cooled-adiabatic architecture. As an intermediate step for higher chip power enablement, OCP ACS group suggested the use of A-L heat exchangers. This enables higher rack density, overall reduced power use and peak performance potential without the need to retrofit facility. This session will discuss the use of rear door heat exchangers as a cold plate enabler for the Olympus server and rack.

Speakers
avatar for Brandon Rubenstein

Brandon Rubenstein

Principal Platform Engineer, Microsoft
I've had the privilege to contribute to Microsoft's success as an engineering director, a design architect and a lead designer. Most recently I've managed the mechanical and power teams responsible for defining, designing, and maintaining Microsoft's cloud business hardware and datacenter... Read More →
avatar for Husam Alissa

Husam Alissa

Sr. Engineer, Microsoft
Husam is a technical lead working for Microsoft Data Center Advanced Development Team with focus on next generation technologies for hardware and infrastructure.
avatar for Robert Lankston

Robert Lankston

Sr. HW Engineer, Microsoft
Robert is a Sr. Mechanical Engineer in Microsoft's Cloud Hardware Engineering team with a focus on thermal design of servers. At Microsoft, Robert contributed to the thermal design and architecture of the Olympus infrastructure as well as many of the server platforms that integrate... Read More →


Tuesday May 12, 2020 1:30pm - 2:00pm
EW: Advanced Cooling Solutions

1:30pm

OAI-UBB OAM Universal Baseboard Design Specification Update
Artificial Intelligence (AI) applications are rapidly evolving and producing an explosion of new types of hardware accelerators for Machine Learning (ML), Deep Learning (DL), and High-Performance Computing (HPC). Different implementations target similar requirements for power/cooling, robustness, serviceability, configuration, programming, management and debug, as well as inter-module communication to scale up and input/output bandwidth to scale out. We need an open infrastructure to intercept rapid innovation in artificial intelligence. OAI is where open accelerator infrastructure meets open artificial intelligence. We contributed the OAM design specification during 2019 OCP Global summit, in this workshop we will present our universal OAM baseboard design proposal from multiple hyperscalers and different OAM suppliers.

Speakers
avatar for Hang Song Kok

Hang Song Kok

Principal Engineer, Intel
avatar for Jose Perez

Jose Perez

Hardware Platform Architect, Intel
I am interested in Server Architecture, Artificial Intelligence Hardware, and Digital Communications.


Tuesday May 12, 2020 1:30pm - 2:00pm
EW: Open Accelerator Infrastructure

1:30pm

Attestation
Servers are filled with a plethora of subsystems, peripherals, accelerators, hardware, firmware, and software products from multiple global suppliers. We present a standardized, open, and automated mechanism to dynamically establish and verify trust in those products through cryptographic attestation.



Speakers
avatar for Elaine Palmer

Elaine Palmer

Senior Technical Staff Member, IBM T.J. Watson Research Center and IBM Academy of Technology
I work on secure (verified) and trusted (measured) boot firmware. My background is in device ownership transfer, hardware security modules, smart cards, FIPS 140-2, and security evaluations.


Tuesday May 12, 2020 1:30pm - 2:15pm
EW: Security

1:30pm

Panel - Learnings from HBM Ecosystem for Optimal Die-to-Die Interconnect Solution
High Bandwidth Memory (HBM) technology has matured over the last few years, and accelerator, GPU and network device provides are shipping products in volumes with the first and the second generations of the HBM memory, using Through Silicon Via (TSV) and 3D packaging.

With the growing interest in multi-chip packaging solutions, either for die disaggregation to split to improve yield and cost; or for the heterogeneous integration by bringing different functions into the same package for improved power, cost, and bandwidth. It is worth to take a closer look at what can be learned from the challenges that were overcome by the different parts of the HBM ecosystem before it was adopted. To find what can be learned, leveraged, and considered, as the ODSA sub-group works on the die-to-die solution specifications.

The panel will bring together experts with different backgrounds and experiences in HBM system design, packaging, test, tools, and end products to share their learnings, in order to address the die-to-die ecosystem needs for broader application and market needs.

Speakers
avatar for Rita Horner

Rita Horner

Product Marketing Manager, Sr. Staff, Synopsys
2.5D/3D IC Technologies with Multi-die integration
avatar for Anand Iyer

Anand Iyer

Director of Product Planning, Samsung Semiconductor, Samsung
CC

Calvin Cheung

Vice President, ASE Group
avatar for Lalitha Immaneni

Lalitha Immaneni

Intel Vice President, Technology Development Group; Director, Advanced Design and Technology Solutions Department, Intel
WG

Wilfred Gomes

Fellow, Intel


Tuesday May 12, 2020 1:30pm - 2:30pm
EW: Open Domain-Specific Architecture

2:00pm

OAM Reference Systems Joint Review
OAI (Open Accelerator Infrastructure) is where open accelerator infrastructure meets open artificial intelligence. Last year, we contributed OAM and universal base board (UBB) design specification with multiple partners. Now we have successfully powered on OAI reference system and will step into a more comprehensive design validation stage in first half of year 2020. In this workshop, Hyve Design Solutions, Inspur, ZT Systems will present reference system updates with power on status as well as demo in the OCP summit.

Speakers
avatar for Tiffany Jin

Tiffany Jin

Mechanical Engineer, Facebook
avatar for Emilie Yang

Emilie Yang

Solution Architect, Hyve Design Solutions
Emilie is a senior Technical Program Manager at Hyve Design Solutions in Taiwan. She contributed several OCP hardware and cooling designs in past decade. Now, she continuously delivers various products through program management and technical expertise to OCP community and custom... Read More →
avatar for YJ Huang

YJ Huang

Development Manager, Inspur
8+ years of relevant experience in electronic products. Major in HW development management. Now is responsible on Inspur OAI reference system development.


Tuesday May 12, 2020 2:00pm - 2:30pm
EW: Open Accelerator Infrastructure

2:15pm

Recovery of Trust
The Recovery principle is a mechanism for restoring Platform Firmware code and critical data to a state of integrity in the event that any such firmware code or critical data are detected to have been corrupted, or when forced to recover through an authorized mechanism. Recovery process of the Platform Active Root-of-Trust (PA-RoT) is a critical process for the overall security of the platform since the PA-RoT will in many cases function as a Root-of-Trust for recovery (RTRec) for symbiotic devices. Guiding principles for this are based on NIST SP 800-193 and the three pillars supporting Platform Resiliency: Protection, Detection, Recovery.

Speakers
avatar for Wojtek Powiertowski

Wojtek Powiertowski

Silicon Security Architect, Facebook
Security Architect working on making silicon more secure and reliable by building Root-of-Trust solutions, mutual authentication and attestation frameworks and always thinking about the recovery scenario because at some point things will break.


Tuesday May 12, 2020 2:15pm - 2:45pm
EW: Security

2:30pm

OAI-SCM - Building on DC-SCM to Support OAI
Demonstrate a concept of OAI-SCM by leveraging DC-SCM. Demonstrate a mockup sample for OCP. Present about the advantages and the benefits of the new architecture and how it promotes modular design concept that fits into existing and new infrastructures for easy adoption.

Speakers
avatar for Siamak Tavallaei

Siamak Tavallaei

Principal Architect, Microsoft Azure; CXL Consortium™ Technical Task Force Co-Chair, Microsoft
Siamak Tavallei is a principal architect and technologist instrumental in development and evolution of innovative multi-node, and multi-processor servers as well as technology initiatives in areas of storage/memory hierarchy and heterogeneous, distributed computing. He is a customer-driven... Read More →
avatar for Raghavan Venugopal

Raghavan Venugopal

Executive Director, Wiwynn
Raghavan Venugopal has greater than 20 years of experience in the computer server industry as an engineer, architect, manager, executive, change agent, and an innovator.He is currently serving as the Executive Director of Customer Engineering and Strategy at Wiwynn for the last 18... Read More →
HH

Huang Hope

Senior Director, Wiwynn


Tuesday May 12, 2020 2:30pm - 3:00pm
EW: Open Accelerator Infrastructure

2:30pm

ODSA: Year in Review
The Open Domain-Specific Architecture is a sub project of the Server project established at the 2019 Global Summit. The ODSA aims to establish both an open ecosystem and an open marketplace of chiplets to support the design of domain-specific accelerators and other devices. In one year, the ODSA has grown to a project with seven active work streams and with active participation from more than 25 companies. To enable the ecosystem and marketplace, the ODSA aims to create an open inter-chiplet interface, provide proof-of-concept demonstration vehicles and establish workflows to use chiplets for product development. This talk will review the progress made by the ODSA, significant milestones, their impact on product development and how new participants can contribute.

Speakers
BV

Bapi Vinnikota

Technologist Architect, OCP


Tuesday May 12, 2020 2:30pm - 3:00pm
EW: Open Domain-Specific Architecture

3:00pm

Lightning Talk: The IEEE HIR and the ODSA
Speakers
avatar for Bill Chen

Bill Chen

ASE Fellow & Senior Technical Advisor, IEEE/ASE
William Chen (Bill) holds the position of ASE Fellow & Senior Technical Advisor at ASE Group. Prior to joining the ASE, he was the Director at the Institute of Materials Research & Engineering (IMRE) in Singapore, following a distinguished career at IBM Corporation. He was a past... Read More →


Tuesday May 12, 2020 3:00pm - 3:05pm
EW: Open Domain-Specific Architecture

3:00pm

Advanced Cooling Solutions for OAI
Open Accelerator Infrastructure (OAI) is a platform built around the Open Accelerator Module (OAM). Last year, we contributed the OAM specification with a reference air-cooled heatsink that could support up to 450W. The module is rated up to 700W at 48V and advanced cooling solutions become necessary for even higher power OAM products. The OAI group is working together to explore different cooling approaches, developing standard cooling solutions for OAM and OAM-based platforms. A universal baseboard (UBB) specification is already available, and we are building upon this with a common tray specification as well. In this talk, we will share the module level and system level liquid cooling solutions we have studied, and the efforts on a universal cold plate design and chassis-level standardization.

Speakers
TG

Tianyi Gao

Engineer, Baidu
CC

Cheng Chen

Engineer, Facebook
avatar for Ethan Cruz

Ethan Cruz

Thermal Engineer, AMD


Tuesday May 12, 2020 3:00pm - 3:30pm
EW: Open Accelerator Infrastructure

3:05pm

Lightning Talk: Heterogeneous Integration – History and Future
Speakers
avatar for Manoj Roge

Manoj Roge

VP, Product Planning & Business Development, Achronix
Member of Executive staff, responsible for driving Product Strategy and Business Development.Driving major architecture innovation and disruption for Data Acceleration workloads!


Tuesday May 12, 2020 3:05pm - 3:10pm
EW: Open Domain-Specific Architecture

3:10pm

3:15pm

Lightning Talk: NXP’S DIPORT DIE TO DIE TECHNOLOGY
Speakers
avatar for Sam Fuller

Sam Fuller

Director of Strategy & Solutions Marketing, Digital Networking, NXP
Innovative and articulate executive leader with more than 20 years’ experience driving semiconductor and electronics business growth. Possesses expertise in embedded processing, networking and Internet of Things (IoT) hardware/software development, and extensive experience working... Read More →


Tuesday May 12, 2020 3:15pm - 3:20pm
EW: Open Domain-Specific Architecture

3:20pm

3:25pm

Lightning Talk: Synopsys and ODSA
Speakers
avatar for Manuel Mota

Manuel Mota

Sr. Product Marketing Manager, Synopsys
Manuel Mota is the product marketing manager responsible for DesignWare Die to Die Interface IP portfolio. He brings more than 18 years of technical and marketing experience to his position, having held similar responsibilities for the Analog and Wireless IP products. Prior to Synopsys... Read More →


Tuesday May 12, 2020 3:25pm - 3:30pm
EW: Open Domain-Specific Architecture

3:30pm

Lightning Talk: Chiplet Design Exchange Flash Update
Speakers
avatar for Jawad Nasrullah

Jawad Nasrullah

CTO, ZGlue
Electronics System DesignChiplets and Heterogeneous Integration


Tuesday May 12, 2020 3:30pm - 3:35pm
EW: Open Domain-Specific Architecture

3:30pm

OAMTool for OAM
The rapidly growing AI applications has created demands for specialized hardware to accelerate their computations and reduce deployment complexity and costs. The OCP community has aligned to the Open Accelerator Module (OAM) form factor for accelerator modules, to ease the design and deployment of these specialized hardware modules.

Aligning the industry to the OAM form factor is the first step to create an interoperable, non-vendor locking Open Accelerator Infrastructure (OAI). Standardizing the management of these OAMs in a vendor-agnostic way is crucial to deploy the OAI in the hyper-scale environments. Towards this goal, we propose to define the OAM management and monitoring tool(OAMTool) . In this talk we will talk OAMTool spec we've collaborated and developed with industry.

Speakers
WZ

Wei Zhang

Software Engineer, Facebook
HK

Harish Kasiviswanathan

Technical Manager, AMD
I am part of Machine Learning and Systems Engineering group within AMD


Tuesday May 12, 2020 3:30pm - 4:00pm
EW: Open Accelerator Infrastructure

3:40pm

3:45pm

Lightning Talk: IBM and ODSA
Speakers
BA

Brian Allison

Senior Technical Staff Engineer, IBM
Brian Allison is an IBM Senior Technical Staff Engineer and Chief Engineer for External Engagements including CAPI and OpenCAPI. Over his 25 year career at IBM, Mr. Allison was the Chief Engineer of various industry leading IBM chip sets that spanned cache coherent, memory and... Read More →


Tuesday May 12, 2020 3:45pm - 4:00pm
EW: Open Domain-Specific Architecture

4:00pm

Baidu AI Infrastructure
Under the strategy of strengthening the mobile foundation and leading in AIî, Baidu has been steadily building an AI ecosystem, with accelerated productization and commercialization. In this talk, we will share Baidu’s AI Infrastructure including AI chip Kunlun, AI Computer X-Man, AI Computing Platform KongMing and deep learning framework PaddlePaddle to support diversified AI workloads including Vision/Speech/NLP/CTR/Recommendation. Especially we will deep dive X-Man’s evolution, and why Baidu initiated OAM/OAI project together with Facebook and Microsoft. We will also share how Baidu successfully solved the 10TB sparse model challenge in CTR/Recommendation with a single Baidu AI Computer X-Man, which is a good show case for OAI Fully Connected topology.

Speakers
avatar for Richard Ding

Richard Ding

Principal AI System Architect, Baidu
Richard Ding is Principal AI System Architect from Baidu Infrastructure Team. He leads architecture design of Baidu AI Computing Platform KongMing including AI computer X-MAN, high-performance parallel file system FAST-F, large scale and distributed communication engine Halo, as well... Read More →


Tuesday May 12, 2020 4:00pm - 4:30pm
EW: Open Accelerator Infrastructure
 
Wednesday, May 13
 

8:00am

LOG IN HERE for the Virtual Summit!
LOG IN HERE: https://onlinexperiences.com/scripts/Server.nxp?LASCmd=L:0&AI=1&InitialDisplay=1&ClientBrowser=0&ShowKey=86206

REGISTER HERE: https://www.opencompute.org/summit/global-summit/registration

Wednesday May 13, 2020 8:00am - 4:30pm

8:59am

OCP Experience Center
OCP Accelerator Module, OAI ( Facebook, Hyve Design Solutions, Inspur, Xilinx)
OAI will demonstrate currently available OAMs as well as show functional implementations in multiple OAI systems.

OpenEDGE (Artesyn Embedded Power, ASUS, BelPower Solutions & Protection, Flex, Inventus Power, MiTAC, NOKIA)
OCP Open Edge project creates open specifications, standards and reference designs for Edge Computing and Telco use cases that require small form factor server designs that enable high performance and low latency computing. Participants of the project are across the ecosystem including Telco Service Providers, OCP solution providers, ODM/OEM manufacturers, and key technology providers. Nokia has contributed the initial Open Edge solution that creates the basis for interoperable designs.

RunBMC (Dropbox, Intel, Vertiv)
This is a demo of hardware modules following OCP RunBMC specification. The first video delivered by authors of the specification from Dropbox features the demo of HS2500 Prototype Module and HS2500 Production Module. This video also presents the standalone RunBMC Bring Up Vehicle and RunBMC-compatible Olympus chassis with Nuvoton Poleg RunBMC Module. The second video delivered by Intel presents FPGA-based RunBMC module featuring Intel® Cyclone V and Intel® MAX10. The video demonstrates the integration of this module in standard Intel 2S Skylake-X platform by using RunBMC Bring Up Vehicle card from Dropbox. In this video you can also see OpenBMC boot on Intel RunBMC module and basic management of the platform with WebUI, SOL and Remote KVM.

Open System Firmware, OSF (Facebook, Wiwynn, HP Enterprise, Intel and ITRenew).
Open system firmware is an open development project, the goal of which is to allow OCP owners to 'own their firmware' -- to move the point of control of firmware to the system owner. Starting in March, 2021, OCP badging for servers will require that systems support OSF. We are exploring multiple ways to accomplish this with components including Hostboot, UEFI, coreboot, and LinuxBoot.

OpenRMC (Inspur)
The Demo will show the establishment of OpenRMC Northbound 1.0 profile execution on top of the bundle of OpenRMC FW & HW (RMC) contributed to OCP as a complete reference solution to address the fulfilment of user's requirements/demand for a comprehensive management of an entire rack system and beyond via RESTful API through remote management portal/client console with detailed WebUI and paving the way for exploring future OpenRMC v2.0.

SEBA Solution (MITAC and EdgeCore)
SEBA solution is one of ONF project and with Edgecore and MiTAC, you can have the OCP hardware ready for SEBA.

OCP NIC 3.0 group (Amphenol, Broadcom, Dell Technologies, Facebook, Intel, Marvell Semiconductor, NVIDIA Networking - Mellanox, Quanta, TE Connectivity, UNH-IOL, Wiwynn Corporation)
The goal of this subgroup is to form a healthy OCP NIC 3.0 ecosystem through a supportive, informative and inclusive community.

96Tb DDC based Disaggregated Router ( Credo, DriveNets, Edgecore Networks, Keysight Technologies, UfiSpace )
The Open Compute Project (OCP) has selected UfiSpace, Credo and DriveNets to give the first ever live demonstration of a Distributed, Disaggregated Chassis (DDC) model as proposed to the OCP by ATT. This demonstration is the largest live demonstration of a distributed router cluster ever given in public. The technology is aligned with the vision of TIP (Telecom Infra Project) who also invited DriveNets and UfiSpace to advise on their Disaggregated Open Router Initiative and ensure its alignment with the OCP's efforts around network disaggregation.

Wednesday May 13, 2020 8:59am - 4:00pm
Experience Center

8:59am

Expo Hall Open
Wednesday May 13, 2020 8:59am - 4:00pm
Expo Hall

8:59am

Expo Hall Talks - ON DEMAND
Wednesday May 13, 2020 8:59am - 4:00pm
Expo Hall Stage

9:00am

OSF Introduction and Transition Schedule
Speakers
avatar for Ron Minnich

Ron Minnich

Software Engineer, Google
linuxboot, u-root, coreboot, linuxbios, ... all open source firmwarelinux kernel, servers,


Wednesday May 13, 2020 9:00am - 9:15am
EW: Open System Firmware

9:00am

Hardware Management Project Overview
Hardware management project provides a common foundation for manageability to a number of OCP projects including servers; storage; data center facility; rack and power. The hardware management project’s scope and charter have been revised and expanded in the past year. This talk will provide an overview of hardware management project and summarize project focus areas including base manageability; rack management; and profiles for OCP projects. The talk will also cover other areas of interest for expansion.

Speakers
avatar for Hemal Shah

Hemal Shah

Distinguished Engineer, Broadcom
Hemal Shah is a Distinguished Engineer and Systems/Software/Standards architect in the Compute and Connectivity (CCX) division at Broadcom Inc. He leads a team of architects. In his role, he is responsible for the product architecture definition of Ethernet NIC products, NIC integration... Read More →


Wednesday May 13, 2020 9:00am - 9:15am
EW: Systems & HW Management

9:00am

An Introduction to ACS Cold Plate Focus Areas
The OCP Advanced Cooling Solution (ACS) Cold Plate work stream are currently working on two contributions/focus areas:
• ACS Liquid Cooling Cold Plate Requirements
• Manifold Design for the Open Rack V3
This presentation gives the status of these different efforts and solicits interest/ feedback from participants to contribute to the current focus areas. The target is to have the ACS Liquid Cooling Cold Plate Requirements approved and published by the Global Summit, while the Manifold Design for the Open Rack V3 is a new focus area for the cold plate workstream.

Speakers
avatar for Jessica Gullbrand

Jessica Gullbrand

Tech Lead Thermal Engineer, Intel
Jessica is a Thermal Tech Lead Engineer at Intel in the Data Platform Group (DPG). She is currently working on energy efficiency, immersion, and liquid cooling cold plate technologies. Jessica has worked on new and innovative technology solutions in fluids, acoustics, and thermals... Read More →


Wednesday May 13, 2020 9:00am - 9:30am
EW: Advanced Cooling Solutions

9:00am

OCP Networking Welcome/Overview
Speakers
avatar for Xin Liu

Xin Liu

Principal Product Manager, Microsoft Azure, Microsoft
avatar for Omar Baldonado

Omar Baldonado

Director of Software Engineering, Infra Network, Facebook
Omar Baldonado is the co-lead of the OCP networking project. At Facebook, he is part of the networking team, responsible for large-scale software systems that model, automate, monitor, and control Facebook’s network. The network domains include Facebook’s data centers; global... Read More →
LY

Lihua Yuan

PARTNER SOFTWARE ENG MGR, Microsoft
Lihua has been with Microsoft for 10 years, working on building hyper scale network since 2008. He got his Ph.D in Computer Science from UC Davis. He has around 11 publications, and a rich experience in speaking on all sorts of conferences, from Sigcomm, to INFOCOM, OCP, ONS etc... Read More →


Wednesday May 13, 2020 9:00am - 9:30am
EW: Networking Hardware

9:00am

DC-SCM Base Specification and Design Details - presented by Microsoft
In this workshop, Wiwynn will co-present with Microsoft on the details of the DC-SCM base specification. DC-SCM is a concept of modular server management that moves common server management, security, and control features from a typical motherboard onto a common form factor module. The idea is to enable a common security and management module form factor and interface which can be used across datacenter platforms. The presentation will include details of the mechanical form factor and electrical interface with example block diagrams providing context for the specification and potential implementations. We will also share design experience from existing DC-SCM development efforts.

Speakers
avatar for Priya Raghu

Priya Raghu

Senior Hardware Engineer, Microsoft
Priya is a Senior Hardware Engineer in Microsoft’s Cloud Hardware Infrastructure Engineering (CHIE) org. Priya has several years of experience designing servers. Her areas of interest include server management, security and power delivery.
avatar for Book Kao

Book Kao

Director, Wiwynn
This is Book from Taiwan Wiwynn, I am the EE lead and have over 20 engineers in my team. Expert in datacenter server HW desiagn and Intel X86 architecture.


Wednesday May 13, 2020 9:00am - 9:30am
Executive Tracks

9:00am

Case Study of Aarki Swapping 100% to OCP+Netris
Case study: Aarki upgrades its data centers in 5 locations with Netris and Mellanox to achieve performance, reliability and agility at a massive scale.

Speakers
avatar for Alex Saroyan

Alex Saroyan

CEO, Netris
CEO of Netris - the new automatic netops platform that turns your physical network into service, and provides network services to your Kubernetes cluster, compute, and storage.


Wednesday May 13, 2020 9:00am - 9:30am
OCP Case Studies

9:15am

Video Transcode Accelerator in Facebook Datacenter
FB is implementing ASIC accelerators in its datacenter to support video transcode workload. In this talk, we will provide a hardware and software overview of the system and components used, including ASIC introduction and module, server configuration. Our Video Transcode platform leverages the widely deployed Yosemite V2 system, where we are able to install 12 ASIC modules that provides a large performance and efficiency win comparing to a standard general-purpose compute server system. Finally, we will provide an overview of the video application and how this accelerator will be used in datacenter.

Speakers
avatar for Hao Shen

Hao Shen

Hardware Engineer, facebook
Hao is hardware engineer at Facebook working on AI platform. His focus is on AI Hardware System, Accelerator enablement and large scale datacenter solution.
JY

Jeremy Yang

Facebook
avatar for Bharath Muthiah

Bharath Muthiah

Technical Sourcing Manager, Facebook
Bharath leads technical sourcing of server technologies for Facebook's Compute, Video, and AI Infrastructure.


Wednesday May 13, 2020 9:15am - 9:30am
EW: Servers

9:15am

OCP OpenRMC Overview and Update
The OpenRMC subproject was established to architect and enabled open-source rack management controllers (RMC). The subproject recently released the OpenRMC northbound interface specification and associated OCP profile. This presentation will present the northbound interface and the subproject’s future direction.

Speakers
avatar for John Leung

John Leung

Principal Engineer, Intel
John Leung is a Principal Engineer at Intel Corporation in the Data Platforms Group. John has been in the computer industry for over 36 years: architecting, designing and coding. At Intel, John is currently focused on the manageability interfaces for data center compute platforms... Read More →
avatar for Han Wang

Han Wang

Principal Architect, Inspur
Dr. Wang is Principal Architect and Senior Client Product Manager at Inspur US. He has been in server, storage, cloud infrastructure industry for more than 16 years and has full experience and solid cross-field expertise in foundational research, engineering development, engineering... Read More →


Wednesday May 13, 2020 9:15am - 9:30am
EW: Systems & HW Management

9:15am

Intel Open Platform Enabling Update
As the Open System Firmware (OSF) workstream progresses towards enabling redistributable firmware images for OCP hardware, there are key dependencies on Intel. This session is designed to communicate Intel’s progress supporting OSF logo efforts.

At the 2019 Global Summit, OSF built a basic roadmap for OSF logo and OSF has been executing to that roadmap. This update at the 2020 Global Summit is key for ODM and OEM customers planning OSF ready systems.

This session includes updates on open source bootloader enabling and redistributable initialization binaries required for the next generation Intel silicon, including the Intel® Firmware Support Package and Intel® Server Platform Services.

Speakers
avatar for Isaac Oram

Isaac Oram

Principal Engineer, Intel
UEFI Firmware for Intel Silicon Enabling


Wednesday May 13, 2020 9:15am - 9:45am
EW: Open System Firmware

9:30am

Accelerator Platform for Inference Applications in Facebook
As footprint of applications utilizing AI and complexity of neural networks used increases , FB is implementing customized HW using ASICs for Machine learning applications.

In this talk, we will discuss the compute platform being developed currently for Inference applications. We will go over the accelerator module form factors (M.2 and dual M.2) and specifications developed at Facebook to support ASICs with different size and power envelopes. The talk will also cover the Hardware platform design (Yosemite V2.50 and Glacier point V2) to support these applications. We will discuss the Glow SW stack that will run on these ASICs and the debug capabilities built into the platform.

The system is developed on the Yosemite 2.50 (Yv2.50) platform which is a modification of the existing Yosemite V2 single socket platform. Yv2.50 is a 2 host system, which supports higher system power and host network bandwidth (50G/host) to improve overall effective system performance. The network cards used have architecture improvements to optimize performance in multihost environments.

The modules are housed in a Glacier point V2 carrier card which are in turn connected to the Twinlake (Intel Skylake) hosts on the Yosemite V2.50 platform through a PCIE switch. Each Glacier point V2 carrier card can house up to 6 dual M.2 accelerator modules or upto 12 M.2 modules.

Glow accepts a computation graph from deep learning frameworks, such as PyTorch, and generates highly optimized code for machine learning accelerators. We will discuss the integration and validation of Glow frameworks on these systems.

Speakers
avatar for Pavan Shetty

Pavan Shetty

Electrical test engineer, Facebook
SP

Soumya Padmanabha

Hardware Systems Engineer, Facebook


Wednesday May 13, 2020 9:30am - 9:45am
EW: Servers

9:30am

Overview of Hardware Fault Management Subproject
System failures due to hardware faults in large scale data center are difficult to diagnose; resulting in inefficient repair action to maintain the quality service of the fleet. Presented here is an overview of the proposed “Hardware Fault Management” subproject within the “System and Hardware Management” working group. The goal of this subproject is to 1) Develop a HW Fault management Solution Guide. 2) Develop Solution Validation Recipes; to assist OCP system and hardware management in large scale data center.

Speakers
ZY

Zhengyu Yang

Hardware System Engineer, Facebook
avatar for Anil Agrawal

Anil Agrawal

Technial Lead (RAS), Intel Corp
Anil has been working as electrical engineer at Intel since 2000. He is passionate about solving customer issues related to system architecture, design, and validation. His focus has been to develop system fault handling solutions. Please refer to Anil\'s linked-in profile for more... Read More →


Wednesday May 13, 2020 9:30am - 9:45am
EW: Systems & HW Management

9:30am

Immersion Requirements Revision 2
This workshop will be focused on the second revision of the Immersion Requirements whitepaper, which is the first qualification step for immersion technology within OCP.

Speakers
avatar for Rolf Brink

Rolf Brink

CEO, Asperitas
Product development for cloud architectures and datacentre infrastructure, combined with a passion for cleantech innovations like liquid cooling, are the foundation of the developments which drive Asperitas.Asperitas is a clean-tech company focused on greening the datacentre industry... Read More →
avatar for Jimil Shah

Jimil Shah

Application Development Engineer, 3M
Immersion Cooling of Data Centers


Wednesday May 13, 2020 9:30am - 10:00am
EW: Advanced Cooling Solutions

9:30am

ONIE Project Update
I'll provide an update on the current state of the Open Network Install Environment, and introduce a new extremely  user friendly build environment to support historical builds and distributed development.  The build enviroment code can be previewed here: https://github.com/ehdoyle/DUE

Speakers
AD

Alex Doyle

ONIE Project Lead / Build engineer, Cumulus Networks
I'm the project lead for the Open Network Install Environment


Wednesday May 13, 2020 9:30am - 10:00am
EW: Networking Hardware

9:30am

DLRM Workloads with Implications on Hardware and System Platforms - presented by Facebook
In this talk we discuss the computation and communication patterns exercised in the deep learning recommendation models (DLRMs). DLRMs are widely deployed at scale for a number business critical applications and exercises all parts of the HW infrastructure – memory, compute, storage and network. In particular, we highlight the differences in terms of the balance between compute, memory and network when compared to the more common computer vision and natural language processing counterparts. We outline different strategies for asynchronous and synchronous distributed training, including the use of model and data parallelism and their impact on system design. The latter requiring high performance interconnects with optimal topology and efficient fabric, supporting Alltoall communication in addition to Allreduce. Finally, we do a deep dive into the open-source implementation of DLRM in Pytorch framework and its use for HW/SW co-design for this important class of applications.

Speakers
avatar for Dheevatsa Mudigere

Dheevatsa Mudigere

Research Scientist, Facebook
- Deep Learning / AI- Scientific Computing- Parallel computing/High performance numerical computing- HW / SW co-design
avatar for Maxim Naumov

Maxim Naumov

Research Scientist, Facebook
Maxim Naumov joined Facebook in January 2018. His interests include deep learning, parallel algorithms and numerical methods. In the past, he held different positions at Nvidia Research, Emerging Applications and Platform teams. He has also worked at Intel Corporation Microprocessor... Read More →


Wednesday May 13, 2020 9:30am - 10:00am
Executive Tracks

9:30am

Ready to Roll, Deploying in OCP Ready™ Colo
Customer case study - deploying a prebuilt OCP server platform for Nexedi's Rapid.Space

Justin will share Hydro66's experience and lessons learned accepting pre-populated OCP racks into a colocation facility designed for traditional 19" equipment, with an emphasis on the practicalities of painless deployment.

The OCP Foundation developed the requirements and process for a colocation data center facility provider to become a listed Colo Solution Provider on the OCP Marketplace. Being an OCP Ready™ Colo Facility means Enterprise clients have an assured way to deploy OCP servers, storage and networking equipment in an extremely cost-effective and scalable way. Compliance means a proven method for accepting OCP pre-built racks, safe in the knowledge that the data center has been pre-vetted and meets or exceeds all the criteria for hosting OCP gear.

Justin will also describe the company's experiences in applying for and gaining OCP Ready™ Facility status. Hydro66 was the first facility in mainland Europe to achieve OCP Ready™ certification.

Speakers
avatar for Justin Fielder

Justin Fielder

CSO, Hydro66
Making OCP compliant rackspace available in an ultra-efficient 100% renewable low cost colocation data center. Enabling you to deploy OCP in a wider eco-system and gain meaningful immediate reductions in operational costs. Take leadership on environmental footprint beyond CSR and... Read More →


Wednesday May 13, 2020 9:30am - 10:00am
OCP Case Studies

9:45am

Improving Cloud Scale Hardware Fault Diagnostics
Traditional Enterprise Class systems have developed and deployed robust hardware fault diagnostics solutions. However; such solutions may not be suitable in Cloud environment. There is a need for scalable hardware fault diagnostics solution for cloud scale deployment. This presentation will cover some of the key learnings from the joint efforts made between Microsoft and Intel in improving Cloud Scale Fault Diagnostics.

Speakers
avatar for Neeraj Ladkani

Neeraj Ladkani

Senior Software Engineer, Microsoft
avatar for Rama Bhimanadhuni

Rama Bhimanadhuni

Principal Firmware Engineer, Microsoft
Firmware, RAS,UEFI, Telemetry


Wednesday May 13, 2020 9:45am - 10:00am
EW: Systems & HW Management

9:45am

Open System Firmware on Arm
This presentation provides an overview of the market reality of the boot system firmware on Arm devices and systems. It covers Arm Trusted Firmware, SBBR (UEFI/ACPI) for the standard OSes and hypervisors, and EBBR (UEFI layer on top of uboot) for embedded devices. It also describes how open source firmware can be achieved on these frameworks. In addition, the session will explore LinuxBoot support on Arm server systems.

Speakers
avatar for Dong Wei

Dong Wei

Fellow, Arm
Dong Wei is a Standards Architect and Fellow. He leads the Arm ServerReady program and its SBSA, SBBR, SBMR and SBSG specs. He is a Board Member on the PCI SIG and CXL Consortium. He is the Chief Executive of the UEFI Forum.
avatar for Samer El-Haj-Mahmoud

Samer El-Haj-Mahmoud

System Architect, Arm
Samer El-Haj-Mahmoud is a Principal Systems Architect at Arm Architecture and Technology Group, working on Arm infrastructure enablement and industry standards. His experience includes server development, firmware, system software, and hardware management. Samer is an active participant... Read More →


Wednesday May 13, 2020 9:45am - 10:10am
EW: Open System Firmware

9:45am

Facebook Next-Gen 2S Server Architecture
Introducing Sonora Pass, Facebook next-generation 2-socket server architecture. It is a base platform which supports new I/O form factors, modular expansion, support ORv2 and ORv3 ready.

In this session, we will be sharing our core feature sets, design flexibility considerations, mechanical architecture, hardware security enhancement and many more.

Speakers
avatar for Damien Chong

Damien Chong

Hardware Engineer, Facebook
avatar for Yin Hang

Yin Hang

Thermal Engineer, Facebook
avatar for Chris Chamberlin

Chris Chamberlin

Mechanical Engineer, Facebook


Wednesday May 13, 2020 9:45am - 10:15am
EW: Servers

10:00am

RunBMC - BMC Standard Interface
RunBMC – BMC interface specification for OCP Baseboard Management Controllers (BMC) perform a range of functions on servers; like checking temperature ranges; voltage; current; power management; and other kinds of health checks. Admins typically monitor these statistics through IPMI; issuing commands through a network topology that’s separate and insulated from the communication between the server’s CPU; firmware; and operating system. There’s a proliferation of hardware devices and SKUs in contemporary data centers; all with different BMCs. Last summer’s controversies over potential hardware vulnerabilities built into server networks have people paying close attention to the BMC subsystem. Having a standardized BMC interface will enable companies to “lock down” their BMC subsystem. Having a design that’s more stable; that rolls at a slower cadence than compute platforms; lets you harden your BMC. It’s really difficult in that environment to control everything and guard against vulnerabilities like man in the middle attacks. By focusing security efforts on a single BMC card; system designers can standardize their own software management stack. The RunBMC specification daughter board card design for use with Open Compute Project motherboards. This specification defines the interface between the Baseboard Management Controller (BMC) subsystem and OCP hardware platforms; such as network or compute motherboards. The daughter board interfaces with hardware platforms through a 260 pin SODIMM DDR4 connector; which is intended for mounting into a mating SODIMM DDR4 socket. The specification defines the IO which should cover the majority of platform BMC requirements.

Speakers
avatar for Eric Shobe

Eric Shobe

Hardware Engineering Manager, Dropbox
avatar for Jared Mednick

Jared Mednick

Hardware Engineer, Dropbox


Wednesday May 13, 2020 10:00am - 10:15am
EW: Systems & HW Management

10:00am

Mixed Environment Hybrid Edge Data Centres
Discuss OPC and HPC in the same facility using different cooling solutions in an edge DC.

Heat Exchangers and how mixed cooling solutions can be successfully deployed in the same environment.

Building up heat output provides a useful waste product,

Speakers
avatar for Don Mitchell

Don Mitchell

Data Center Division Manager, Victaulic
Key topics: Liquid distribution & cooling solutions; density diversity; energy reuse; digital streamlining of data center constructionBackground - 20+ years in mission critical design, construction, operation, both data centers, nuclear submarines, missile systems
avatar for Tim Chambers

Tim Chambers

CEO, coolDC
Tim Chambers is Founder and CEO of coolDC. With over 30 years’ experience of working in the data centre and allied industries, he is no stranger to the challenges it faces.Tim has been vocal in his frustration concerning the industry’s reluctance to depart from the practice of... Read More →


Wednesday May 13, 2020 10:00am - 10:30am
EW: Advanced Cooling Solutions

10:00am

Open Networking with Arista
Arista has a long history of collaboration & openness while maintaining high standards of quality & resiliency for its products. This session covers Arista’s learning through the co-development process of Hardware & Software as well as supporting large cloud scale networks with Open Solutions. We'll talk about our experience in driving SAI Macsec this past year, from gathering inspiration from code reuse in our co-development projects to our experiences in the logistics of contributing to SAI.

Join Arista’s Systems Engineer, Narayanan Suryanarayanan and Software Engineer, Ryan Torres as they walk through Arista’s commitment to SONiC, SAI & SAI Macsec. If you're thinking about taking a larger role in SAI, come and join this talk.

Speakers
avatar for Narayanan Suryanarayanan

Narayanan Suryanarayanan

Systems Engineering, Arista Networks
avatar for Ryan Torres

Ryan Torres

Software Engineering, Arista Networks


Wednesday May 13, 2020 10:00am - 10:30am
EW: Networking Hardware

10:00am

Continued Intel Innovation in the Cloud and Edge - presented by Intel
Highlight challenges and potential solutions for managing cloud at scale.  In this session you’ll see how the Cloud and OCP requires not just hardware but solutions. You’ll get insights from topics such as: OCP Hardware Management Project, Intel® Firmware Support Package and Coreboot, liquid cooling solutions, Cloud Edge Server, and hear from a special guest speaker about Power Management Innovations.

Speakers
avatar for Mohan J Kumar

Mohan J Kumar

Fellow, Intel
Mohan Kumar is an Intel Fellow in the Data Platforms Group and the director of Cloud Architecture and Solutions in the Cloud & Enterprise Solutions Group at Intel Corporation. His organization is responsible for driving the architecture requirements for Intel Cloud customers.Kumar... Read More →


Wednesday May 13, 2020 10:00am - 10:30am
Executive Tracks

10:00am

From Servers to Serverless in Ten Minutes
The goal of this session is to overview, outline, and explain the technology to deploy & provision a scale-out kubernetes cluster for running & orchestrating containers & VMs on bare metal. Much of the kubernetes ecosystem is focused on cloud & shared infrastructure platforms, but there are certain customer use cases & characteristics that simply require bare-metal nodes - in fact, even cloud platforms continue to run on bare-metal servers one way or another ;)

This presentation will present our experience taking a high-density OCP rack with over 1,000 compute cores and 9TB of memory from bare hardware to running a serverless demo app in 10 minutes. We will outline the choice of OS components, network connectivity, deployment of master & shared service nodes, and parallel deployment of compute nodes, preparing everything to process application traffic.

There are many layers to the kubernetes deployment stack, including infrastructure components built on other, lower level infrastructure components - sometimes four and five and six layers deep. This provides convenient abstractions for different engineering, deployment, and service levels - but also results in a large amount of total deployed software when considered end-to-end. Deploying such clusters on racks of bare metal gear - whether OCP or otherwise - requires preparation and planning, but does not have to be a months-long endeavor. In order to support customers and SPs that require scale-out on-premise hardware solutions, we have had to evaluate and optimize various approaches to bare metal deployment and will share our learnings here.

Speakers
avatar for Erik Riedel

Erik Riedel

SVP of Compute & Storage Solutions, ITRenew/Sesame
Long-time engineer & engineering leader in infrastructure & storage - interested in bare metal k8s, hardware, and performance at scale.


Wednesday May 13, 2020 10:00am - 10:30am
OCP Case Studies

10:10am

Open Firmware on ProLiant platforms
Discuss HPE’s journey to enable open source firmware on the ProLiant product line. Multiple industry customers and leaders, to be named later, will be involved in the discussion to share the benefits to the industry of such OEM enablement as well as the lowering risks of transitioning with increased maturity of open source projects.

Speakers
avatar for Jeoff Krontz

Jeoff Krontz

Sr. Director, Hyperscale Engineering, HPE
Jeoff Krontz is a customer-facing engineering leader with a focus on delivering innovative high-quality products and solutions to meet customer and market demand. With a deep foundation in product design, technology trends, and manufacturing practices, Jeoff is passionate about next-generation... Read More →
avatar for Jean Marie Verdun

Jean Marie Verdun

Open Platform Technologist, HPE
Jean-Marie Verdun, Senior Strategist Open Platforms at HPE where he works with the Advanced Engineering team to define and execute Open Platforms strategy, with a primary focus on implementing Open Source Firmware on HPE upcoming hardware platforms. Before joining HPE, Jean-Marie... Read More →


Wednesday May 13, 2020 10:10am - 10:35am
EW: Open System Firmware

10:15am

Intel CycloneV-Based RunBMC Solution with OpenBMC Firmware
OCP RunBMC definition provides a standard connector definition that allows for separating the cadence of the platform being managed and the Management/Security Hardware (e.g. PRoT and BMC). The benefit of this approach allows for focusing security efforts on a single BMC card that can be used in a variety of platforms. Intel has a strong commitment to OpenBMC and is actively developing and contributing to OpenBMC.  One of the key benefits that Intel sees in RunBMC is that in early platform development; the BMC silicon is often not ready for initial bring-up. By leveraging RunBMC, we are able to remove this dependency and risk. In addition to RunBMC providing customers choice in the BMC eco-system; it allows for innovation around new implementations of BMC and Security Hardware. To this end Intel has developed an FPGA Based RunBMC card leverage Intel® Cyclone V and Intel® MAX10 . This implementation runs an OpenBMC Firmware Stack. The benefits of using an FPGA based approach; allows for developing Robust RTL but also allows for more rapid innovation of new management and security features which are critical in improving the robustness of BMC and Security Technologies.  By following OCP standards such as RunBMC and OpenBMC this design echoes the OCP mission and vision and provides an open; scalable and flexible platform that enhance possibilities for innovations around BMC in the FW and HW.

Speakers
avatar for Johan van de Groenendaal

Johan van de Groenendaal

Senior PE, Intel
Johan is an expert in DataCenter Management across Cloud and Enterprise at all levels of the Management Stack covering both Software, Firmware and Silicon. His primary responsibilities at Intel is driving the architecture of Intel's @scale Telemetry and Debug Technologies.
avatar for Kasper Wszolek

Kasper Wszolek

Software Architect, Intel
Kasper is a Software Architect focused on manageability firmware solutions for servers. He's been working on the implementation and design of multiple generations of Intel SPS ME FW. His main area of expertise is related to embedded operating systems internals, performance/resource... Read More →
avatar for Qian Wang

Qian Wang

Software Architect, Intel
Qian is an engineer, who has worked in enterprise server BIOS, BMC and hardware design. Her current area of focus is datacenter, server platform manageability architecture and pathfinding.


Wednesday May 13, 2020 10:15am - 10:30am
EW: Systems & HW Management

10:15am

DC-SCM Overview and Benefits
In this workshop, Google will copresent with Microsoft a concept of modular server management and security that moves common server management, security, and control features from a typical processor motherboard onto a smaller common form factor module.  The idea is to enable a common security and management module form factor and interface which can be used across datacenter platforms.  We will introduce the concept and discuss the architectural advances across the development and product lifecycle of the server.

Speakers
avatar for Mark A. Shaw

Mark A. Shaw

Principal HW Eng Mgr, Microsoft
PC

Prakash Chauhan

Server Systems Architect, Google


Wednesday May 13, 2020 10:15am - 10:35am
EW: Servers

10:30am

OpenBMC and Redfish
Redfish is an open industry standard specification for hardware management. It is defined by the DMTF; Distributed Management Task Force. The Redfish Forum was formed in 2014 with the first specification coming in August 2015. New specifications are released every 3-4 months with the latest being 2019.4. Redfish is a replacement for IPMI-over-LAN; a 21-year-old standard. OpenBMC is an open-source Baseboard Management Controller (BMC) firmware stack. In March 2018; OpenBMC became a Linux Foundation project. IBM; Intel; and other OpenBMC contributors are working closely to implement Redfish on OpenBMC. The OpenBMC 2.7 release includes Redfish support for firmware update; inventory; sensors; user management; and event logs. The OpenBMC 2.8 release; scheduled for May 2020; will include Redfish support for LDAP; certificates; telemetry; and events. OpenBMC will continue to add Redfish schemas and properties and is working with the Redfish Forum to add schemas and properties not currently in the Redfish API. More information on OpenBMC's implementation can be found at https://github.com/openbmc/bmcweb.

Speakers
avatar for Gunnar Mills

Gunnar Mills

Software Engineer working on the OpenBMC project, IBM
Gunnar is a Software Engineer at IBM working on the OpenBMC Project. He is maintainer of 4 OpenBMC repositories, including the Web Interface, and the IBM lead for the Redfish on OpenBMC deliverable.


Wednesday May 13, 2020 10:30am - 10:45am
EW: Systems & HW Management

10:30am

Quick Disconnects for Liquid Cooling: Design, Specification & Implementation
With the growing utilization of liquid cooling to address the high temperature output of electronics and computing systems, the impact of individual components on flow rate is often misunderstood or undervalued. As designers embrace the fact that liquid cooling is better at transferring heat at much lower mass flow rates compared to air cooling, attention to thermodynamics in combination with flow rate is critical -- whether the liquid cooling application is for HPC, electric vehicle charging, data centers or laser diode cooling.

This presentation will address individual component selection that affect flow rates and pressures and why variability matters. Discussion will include pumps and system pressure, air inclusion, coolant selection, QD valve design, terminations, circuit tubing diameter, potential failure scenarios and testing for risk mitigation. It will also include a deep dive into quick disconnects. You’ll learn why/how the selection of coolant is dependent upon the application and how this impacts the material selection requirements for each subsequent component in the liquid cooling system. Attendees will leave with an understanding of how to calculate the total pressure drop in the liquid line (via expansions, contractions, bends, valves), what is not often understood about quick disconnects, why pump sizing matters in fulfilling hydrodynamic requirements to address thermal requirements and what to look for in designing an optimized system.

Speakers
avatar for Elizabeth Langer

Elizabeth Langer

R&D Engineering Manager - Thermal, CPC
Beth Langer is the Lead Technical Engineer in the Thermal Management Business Unit at CPC where all connectors manufactured for liquid cooling applications meet or exceed established criterion.


Wednesday May 13, 2020 10:30am - 11:00am
EW: Advanced Cooling Solutions

10:30am

WHY are you still using FLASH like a drive? - presented by KIOXIA
The Hyperscale is redefining storage and virtualization to best orchestrate digital storage for cloud applications in a drive for greater efficiency at scale.  Effectively and efficiently utilizing flash for stable and predictable latency in ever-changing cloud workloads through direct host management of flash is critical to achieving this goal.  

KIOXIA is announcing a technology that redefines digital storage into a new approach that joins software flexibility, host control and flash native semantics into a “Flash Native API” and purpose-built hardware that maximizes the value of flash memory for cloud providers. This combination of technologies fundamentally redefines the relationship between the host and solid-state storage devices and bypasses legacy HDD storage paradigms to unlock host control and enable the use of flash at its natural speed. This approach provides a path-way to flash native architectures that can include traditional block, ZNS or custom FTL implementations and much more.

The “Software Enabled Flash” API and purpose-built hardware provides storage systems architects significant benefits that include ease of adoption and abstraction of flash generational changes. We will show you how we achieve this through flexible open source API’s, libraries and reference designs that insulate the implementer from low-level media management.


Speakers
avatar for Eric Ries

Eric Ries

SVP Memory and Storage Strategy Division, Kioxia America Inc
Eric Ries joined Toshiba Corp. in June 2013 as the General Manager of Corporate Development and Strategy for the Semiconductor & Storage Products Company in Tokyo.  He  returned to the United States in 2016 as the Sr. Vice President for Memory and Storage Strategy for Toshiba Memory... Read More →
avatar for Rory Bolt

Rory Bolt

Principal Architect, Sr. Fellow, Kioxia America, Inc.
Rory joined Kioxia American in 2017.  He has founded, built teams, and delivered product at 4 storage startups which were acquired ($400M, $165M and two undisclosed). Rory has more than twenty five years of experience in data storage systems, data protection systems, and high performance... Read More →


Wednesday May 13, 2020 10:30am - 11:00am
Executive Tracks

10:30am

Profiling OCP Servers for Holistic Air Cooling
Minimizing data center energy use requires a 'holistic' cooling approach, implying that both the air handler and server fans must be operated with control algorithms that work together. While some conditions can increase efficiencies of the cooling system, the effect on the server fans and CPU current leakages must also be considered. To obtain the recipe to run the data center at maximum efficiency a 'sweet spot' is identified, yielding conditions that are favorable for both the cooling equipment and servers.

OCP servers are tested in a windtunnel at the ICE data center research facility at RISE Research Institutes of Sweden in LuleÂ. The 10 meter windtunnel consits of a heat exchanger for cooling, a heater, humidifier and a fan. The windtunnel can be used to control the pressure, temperature and humidity at the inlet to the server while measuring parameters such as temperature differential and server airflow.

The workshop presents a custom fan controller that has been implemented on OCP servers to maintain CPU temperatures at a desired setpoint irrespective of the workload, inlet temperature or pressure. The cooling control software is adjusted depending on operating conditions of the servers to minimize the cooling energy.

Adopting this holistic cooling approach with OCP servers and EcoCooling direct fresh air ventilation units has achieved PUE values less than 1.01 as demonstrated in the EU Horizon 2020 funded project, Boden Type DC, where two 150kW OCP data centers were tuned with holistic control tactics.

Speakers
avatar for Jon Summers

Jon Summers

Scientific Leader in Data Centers, Research Institutes of Sweden
My main area of research these days is on thermal and energy management of the digital infrastructure. I have worked in a research environment focused on heat transfer from the chip to the chiller in the backdrop of using real IT equipment, which lead me to join RISE in the north... Read More →
avatar for Jeffrey Sarkinen

Jeffrey Sarkinen

Research Engineer, RISE Research Institutes of Sweden
My area of interest is data centers with a focus on energy efficiency. Being involved in research projects has allowed to gain experience in data collection and monitoring, automatic control systems, data analysis and thermal management. I studied electrical engineering at the University... Read More →


Wednesday May 13, 2020 10:30am - 11:00am
OCP Case Studies

10:30am

Panel: Network Operating Systems, Disaggregation, and OCP Networking
In this panel, representatives from NOS providers will share their perspectives on disaggregation, including how OCP has enabled their NOS development as well as what the OCP networking ecosystem could be doing to better help NOS providers.

Speakers
avatar for Xin Liu

Xin Liu

Principal Product Manager, Microsoft Azure, Microsoft
avatar for Joe Whitehouse

Joe Whitehouse

Networking Software, Metaswitch
Joe Whitehouse has worldwide sales and busdev responsibility for Metaswitch’s suite of networking software solutions.
avatar for Omar Baldonado

Omar Baldonado

Director of Software Engineering, Infra Network, Facebook
Omar Baldonado is the co-lead of the OCP networking project. At Facebook, he is part of the networking team, responsible for large-scale software systems that model, automate, monitor, and control Facebook’s network. The network domains include Facebook’s data centers; global... Read More →
avatar for Ricardo Pianta

Ricardo Pianta

Engineering Manager, Datacom - Teracom Telematica
Besides his main responsability leading the R&D team at Datacom, Ricardo enjoys his business development role and represents the company in a number of foruns like ONF, OCP and TIP. He became a huge enthusiat of disaggregation initiatives and a very active member of the community... Read More →
avatar for Pete Lumbis

Pete Lumbis

Director of Technical Marketing, Cumulus
Pete is in charge of technical marketing and user documentation at Cumulus Networks. Pete helps customers design, build and automate next-generation datacenter and campus networks. Before Cumulus Networks Pete was a member of the global routing escalation team within the Cisco TAC... Read More →
avatar for Philippe Michelet

Philippe Michelet

Vice President of Product Management, Kaloom
25 years of experience in networking, data center, virtualization, Network OS. Kaloom delivers a highly differentiated \\\"Edge\\\" environment leveraging programmable HW (ASIC/FPGA with P4), native RH/OpenShift OS with our containers executing control plane functions, in the context... Read More →


Wednesday May 13, 2020 10:30am - 11:30am
EW: Networking Hardware

10:35am

Coreboot/linuxboot Feature Development for Server
This joint presentation between Wiwynn and Facebook gives an update on coreboot/linuxboot feature development for server. It discusses status/design of some server features needed by server. The plans and challenges are also discussed.

Speakers
avatar for Jonathan Zhang

Jonathan Zhang

Software Engineer, Facebook
Jonathan is passionate at architecting lower level system firmware/software to achieve computer system design goals. He used to work at software organization of hardware companies (silicon vendors), he is working at hardware organization of software company (hyper-scaler). He is thrilled... Read More →
avatar for Johnny Lin

Johnny Lin

Software Engineer, Wiwynn
I had been working on Linux embedded system bootloader, kernel driver and user space software. My current focus is on coreboot and Linuxboot projects for server.


Wednesday May 13, 2020 10:35am - 10:55am
EW: Open System Firmware

10:35am

DC-SCM Base Specification and Design Details
In this workshop, Wiwynn will copresent with Microsoft details of the DC-SCM base specification.  DC-SCM is a concept of modular server management that moves common server management, security, and control features from a typical motherboard onto a common form factor module.  The idea is to enable a common security and management module form factor and interface which can be used across datacenter platforms. The presentation will include details of the mechanical form factor and electrical interface with example block diagrams providing context for the specification and potential implementations.  We will also share design experience from existing DC-SCM development efforts.

Speakers
avatar for Priya Raghu

Priya Raghu

Senior Hardware Engineer, Microsoft
Priya is a Senior Hardware Engineer in Microsoft’s Cloud Hardware Infrastructure Engineering (CHIE) org. Priya has several years of experience designing servers. Her areas of interest include server management, security and power delivery.
avatar for Book Kao

Book Kao

Director, Wiwynn
This is Book from Taiwan Wiwynn, I am the EE lead and have over 20 engineers in my team. Expert in datacenter server HW desiagn and Intel X86 architecture.


Wednesday May 13, 2020 10:35am - 10:55am
EW: Servers

10:45am

System Manageability for a Very Modular Design
Yosemite platform is designed for datacenter compute and accelerator services. The platform design comprises of a sled with 2 to 4 individual servers in it. The servers can be configured with multi-host NIC or single-host NIC. The system configuration could vary based on expansion cards either directly attached to server board (1OU) or through a riser card to the server board (2OU). Each of the server, NIC and expansion card in the sled has devices that 3 major function that needs to be addressed in-band or/and out-of-band 1. Detection status 2. Firmware upgrade 3. Telemetry 4. Failure detection To manage the modularity Board Management system is architected to have a Board Management Chip (BMC) associated with NIC to enable network based management and each board has a Bridge IC (BIC).

Yosemite platform will make use of OpenBMC as its out-of-band management software. OpenBMC is completely open sourced and is customizable by the end user. OpenBMC treats each system component (NIC, server, expansion and riser cards) as Field Replaceable Unit (FRUs) and would auto detect FRU presence on power up and configurable its management feature per current system configuration. Once system configuration has been determined OpenBMC would launch health monitoring daemons to monitor each FRU’s health status and provide a list of command line utilities for user to perform firmware upgrade on applicable components within each FRU.

Speakers
BW

Ben Wei

System Software Engineer, Facebook
avatar for Deepa Saxena Mehta

Deepa Saxena Mehta

Hardware Systems Engineer, Facebook
A life-long problem solver, I have thrived on different technological challenges through my 20 years career, from memory devices, DIMM design, to client and server platforms. Each technology and platform highlights new issues and new oppurutnity. In my current role in Facebook as... Read More →


Wednesday May 13, 2020 10:45am - 11:00am
EW: Systems & HW Management

10:55am

Future Inference Accelerator Module Form Factor
To fulfill the needs of scaling FB infrastructure and improving performance, power, and efficiency, the development of specialized accelerators and holistic system-level solutions has shown great importance. In order to gain efficiency, thermal/mechanical design and management for those high density, small form factor modules is an important building block to ensure the overall efficiency of the solution. This talk will discuss mechanical and thermal considerations of both current and future generation of the platform for this particular use cases, provide opportunities for further efficiency gain and seek for future collaboration on standardizing and scaling the solution. The talk will also expand and share challenges of the current form factor (M.2 and dual M.2) and discuss further potential scalable form factors.

Topics that might be covered during this talk:
1. FB Glacier Point V2 architecture in YosemiteV2 platform
2. FB next generation architecture (code name TBD) in next generation platform
3. Comparison between the system generation and potential thermal efficiency and power efficiency gain
4. Challenges and Call out for Action:
Form factor impact from die size, heatsink size, orientation, new cooling technologies perspective
Potential intersection with PCIe Gen5
Package design vs. system design - how can we work together as a community

Speakers
avatar for Yueming Li

Yueming Li

Thermal Engineer, Facebook
MH

Michael Haken

Mechanical Engineer, Facebook
avatar for Hao Shen

Hao Shen

Hardware Engineer, facebook
Hao is hardware engineer at Facebook working on AI platform. His focus is on AI Hardware System, Accelerator enablement and large scale datacenter solution.


Wednesday May 13, 2020 10:55am - 11:10am
EW: Servers

10:55am

Vendor Panel
During this panel discussion, various speakers from the morning will weigh into the unique challenges of Open System Firmware.

Ron Minnich is moderating and driving the discussion. Questions will also be open to the floor.
The focus is on the OSF 2021 Transition Schedule with topics including (but not limited to) redistribution of binary blobs, firmware ownership and reusability, and documentation.

Moderators
avatar for Ron Minnich

Ron Minnich

Software Engineer, Google
linuxboot, u-root, coreboot, linuxbios, ... all open source firmwarelinux kernel, servers,

Speakers
avatar for Jean Marie Verdun

Jean Marie Verdun

Open Platform Technologist, HPE
Jean-Marie Verdun, Senior Strategist Open Platforms at HPE where he works with the Advanced Engineering team to define and execute Open Platforms strategy, with a primary focus on implementing Open Source Firmware on HPE upcoming hardware platforms. Before joining HPE, Jean-Marie... Read More →
avatar for Jeoff Krontz

Jeoff Krontz

Sr. Director, Hyperscale Engineering, HPE
Jeoff Krontz is a customer-facing engineering leader with a focus on delivering innovative high-quality products and solutions to meet customer and market demand. With a deep foundation in product design, technology trends, and manufacturing practices, Jeoff is passionate about next-generation... Read More →
avatar for Elaine Palmer

Elaine Palmer

Senior Technical Staff Member, IBM T.J. Watson Research Center and IBM Academy of Technology
I work on secure (verified) and trusted (measured) boot firmware. My background is in device ownership transfer, hardware security modules, smart cards, FIPS 140-2, and security evaluations.
avatar for Dong Wei

Dong Wei

Fellow, Arm
Dong Wei is a Standards Architect and Fellow. He leads the Arm ServerReady program and its SBSA, SBBR, SBMR and SBSG specs. He is a Board Member on the PCI SIG and CXL Consortium. He is the Chief Executive of the UEFI Forum.
avatar for Isaac Oram

Isaac Oram

Principal Engineer, Intel
UEFI Firmware for Intel Silicon Enabling


Wednesday May 13, 2020 10:55am - 11:50am
EW: Open System Firmware

11:00am

OpenBMC in Production: Twitter's Collaborative Journey
Twitter has been collaborating with industry partners for the past 15 months to deploy OpenBMC in our production datacenters. In this talk we detail our motivation for considering OpenBMC; our efforts to bring together partners to collaborate in a common codebase; issues and obstacles faced; and finally our progress and production readiness. We’ll discuss how we worked with peer companies to identify a common feature set; the enhancements that we needed for our adoption; and how we bridged the gap between the incumbent commerical solution and OpenBMC. This presentation is targeted towards organizations that experience a high level of BMC software related issues and want to take more control of their own solution.

Speakers
avatar for Matthew Singer

Matthew Singer

Sr. Staff Hardware Engineer, Twitter
Matt Singer is Senior Staff Hardware Engineer at Twitter. An expert in performance and server architecture, Matt is the technical leader for the hardware team that created one of the most resilient and cost-effective cloud infrastructures in the industry. He is also the architect... Read More →
avatar for Varun Sampat

Varun Sampat

Staff Hardware Systems Engineer, Twitter
Varun Sampat is a Staff Hardware Systems Engineer at Twitter. He has over 12 years experience in system software development for datacenter servers, with a focus on embedded BMC controller software development and new server bring-up. Varun currently leads the OpenBMC software development... Read More →


Wednesday May 13, 2020 11:00am - 11:15am
EW: Systems & HW Management

11:00am

Takeaways After Two Years of Immersion
Since 2018, 2CRSi is accumulating experience in designing servers for immersion. In what follows we would like to share what we have learned during this period. The presentation will be divided in 4 main chapters:
- Material compatibility: here we will speak about the interactivity between different components and the immersion liquid. What works and what should be avoided.
- Mechanical: this chapter in focused on the design features that should be considered when creating a server for immersion.
- Electrical: in this chapter we will cover some aspects related to electrical components interaction with the immersion liquid
- Software: mostly related to PSU’s software adaptation.

Speakers
avatar for Michaël Helezen

Michaël Helezen

R&D engineer, 2CRSI


Wednesday May 13, 2020 11:00am - 11:30am
EW: Advanced Cooling Solutions

11:00am

Open from Cloud to Edge, Building Our Solutions with Modular Design - presented by Wiwynn
While Wiwynn has abundant cloud product and technologies expertise, we like to build more flexible and easy managed platforms with modular design from cloud to edge infrastructure.

There are multiple features of modular designed building blocks we considered, including:
1. Processing power from Intel, AMD to ARM
2. Flexible I/O capability 3. Space fitting constraints 4. Remote deployment & management with diagnosis and self-healing
5. Support different form factors for performance accelerators
6. Efficient cooling and power design

We will show our products and technologies based on modular design. By those features, we can enable workload optimized product design with flexibility, availability, adaptability, and serviceability.

Speakers
avatar for Raghavan Venugopal

Raghavan Venugopal

Executive Director, Wiwynn
Raghavan Venugopal has greater than 20 years of experience in the computer server industry as an engineer, architect, manager, executive, change agent, and an innovator.He is currently serving as the Executive Director of Customer Engineering and Strategy at Wiwynn for the last 18... Read More →


Wednesday May 13, 2020 11:00am - 11:30am
Executive Tracks

11:00am

BTDC: A Practical Demonstration of World Class Efficiency
The Boden Type Data Center (BTDC) is an EU funded Horizon 2020 project with the key objective of demonstrating a range of technologies, creating a small scale OCP style data center that operates with extreme efficiency. This 500kW project includes a 150kW pod dedicated to the efficient operation of legacy Windmill OCP equipment. Additionally, a 150kW module is located in the ICE data center research facility at RISE Research Institutes of Sweden in LuleÂ, which houses Winterfell OCP equipment. The data centers are both equipped with EcoCooling direct fresh air ventilation units that contain supplemental evaporative function to provide cooling. This case study involves a live demonstration of the two data centers to show how new holistic process controls can be employed to minimize the energy used in providing an appropriate thermal environment for the servers and maintaining stable microprocessor operating conditions. One data center shows such stability under fluctuating workloads. The other data center highlights similar stability of microprocessor temperatures under varying supply temperatures, which reflects different climates. In both case the effectiveness of the cooling is exemplified by achieving PUE values of less than 1.01.

The workshop explains how such a performance can be achieved using OCP equipment with only software changes and simple inter-connectivity of the microprocessors, a custom server fan speed algorithm and the cooler control system. Consideration is given as to whether the demonstrated integrated approach could give OCP hardware a competitive advantage in emerging markets.

Speakers
avatar for Jon Summers

Jon Summers

Scientific Leader in Data Centers, Research Institutes of Sweden
My main area of research these days is on thermal and energy management of the digital infrastructure. I have worked in a research environment focused on heat transfer from the chip to the chiller in the backdrop of using real IT equipment, which lead me to join RISE in the north... Read More →
avatar for Jeffrey Sarkinen

Jeffrey Sarkinen

Research Engineer, RISE Research Institutes of Sweden
My area of interest is data centers with a focus on energy efficiency. Being involved in research projects has allowed to gain experience in data collection and monitoring, automatic control systems, data analysis and thermal management. I studied electrical engineering at the University... Read More →


Wednesday May 13, 2020 11:00am - 11:30am
OCP Case Studies

11:10am

Yosemite-V3; the New Platform for Single Socket Server
With evolving use cases, we have taken a more modular approach to designing the single socket server platform. The new versatile platform supports the general compute, storage and hardware accelerator services. The platform comprises of a sled that can support up to 4 single socket 1U server blades with a common Baseboard for management. Each blade could be independently serviced. 3 of those sleds make up the chassis. 8 chassis can be accommodated in an open rack (ORV2).

For HW accelerator and high bandwidth storage applications, the sled could be populated with 2 server blades creating room for those modules. The accelerator and storage modules connect to server through expansion cards. For the network interface, depending on the application, the servers can have a dedicated single host NIC or a shared multi-host NIC.

Speakers
KV

Kiran Vemuri

Hardware Engineer, Facebook
Kiran Vemuri is a Hardware Engineer in the Infra Foundation team at  Facebook, leading the high density, single socket server platforms. He has over 20 years experience building hardware for Enterprise, Consumer and Automotive systems. Prior to Facebook, he worked at Tesla where... Read More →
RE

Ryan Emmett

Technical Program Manager, Facebook
Ryan Emmett is a Technical Program Manager in the Infra Foundation team at Facebook, managing the high density, single socket server platforms. He has over 14 years’ experience building hardware and systems for Cloud and Enterprise server and networking systems. Prior to Facebook... Read More →


Wednesday May 13, 2020 11:10am - 11:25am
EW: Servers

11:15am

Delivering Telemetry @ Scale with Intel and Splunk
The volume of telemetry that a single platform can generate out-of-band continues to grow and with a strong push to be able to do more monitoring out-of-band is requiring new approaches on how telemetry is collected and sent to the Management Infrastructure. Currently BMC solution are generally implemented as poll-based solutions using IPMI or RedFish APIs. This approach generally is restrictive and does not scale well in Hyper-Data Centers. Intel has developed an OpenBMC plugin that consists of a Redfish based API for configuration with a streaming plugin that is capable of sending Intel Autonomous Crash Dump, BMC SelEvents and various forms of Telemetry from the BMC to SPLUNK or Kafka.

In this presentation we will share the details of the architecture and demonstrate how aggregating data from the BMC with IT operations capabilities represented by SPLUNK delivers an @Scale solution for improving overall management of the datacenter.

Speakers
avatar for Mariusz Oriol

Mariusz Oriol

Principal Engineer, Intel
avatar for Johan van de Groenendaal

Johan van de Groenendaal

Senior PE, Intel
Johan is an expert in DataCenter Management across Cloud and Enterprise at all levels of the Management Stack covering both Software, Firmware and Silicon. His primary responsibilities at Intel is driving the architecture of Intel's @scale Telemetry and Debug Technologies.
avatar for Jason Conger

Jason Conger

Architect, Splunk


Wednesday May 13, 2020 11:15am - 11:30am
EW: Systems & HW Management

11:25am

Delta Lake 1S Cedar Island Platform
The Delta Lake 1S server utilizes the performance and advanced intelligence of Intel® Xeon® processors. Delta Lake is engineered to perform across a range of functions, including as a powerful web tier engine as well as a flexible and efficient head node for storage and accelerator applications.

Speakers
avatar for John Bryan

John Bryan

Platform Architect, Intel
John Bryan is a Platform Architect at Intel Corporation, working on high-density servers solutions for the data center. John has collaborated with Facebook on the Mono Lake, Twin Lakes, and Delta Lake one-socket servers based on Intel Xeon SOCs. John has designed servers and supercomputers... Read More →
TW

Todd Westhauser

Hardware Engineer, Facebook
Todd is a hardware engineer at Facebook and the lead hardware engineer on Facebook's next generation 1 socket server, Delta Lake. In addition to server design, Todd has also worked on PCIe Add-In cards, SSD's, and the flight computer for small satellites.


Wednesday May 13, 2020 11:25am - 11:40am
EW: Servers

11:30am

Managing Componentized Servers Using Redfish
Accelerator hardware and pluggable modules are getting more advanced; and managing this hardware often requires increasingly custom solutions. However; when creating new servers it would be simpler to separate swappable components. Fortunately; BMC hardware has become cheaper and more powerful. This makes it possible to split a complex server into multiple management domains; each with its own dedicated Redfish service running on Open BMC. Redfish is a modern management protocol that exposes a RESTful API over HTTP. This talk will explore the design of these Redfish services; and how to aggregate these individual Redfish services into a single logical service. It should be interesting to people in the Open BMC community; as well as anyone interested in Redfish.

Speakers
avatar for Nancy Yuen

Nancy Yuen

Senior Software Engineer, Google
Nancy Yuen is a Software Engineer at Google. She has been writing software for 20 years, mostly kernel and system software for embedded optical switches and new server bringup. She has worked on several server generations deployed in Google's fleet. She also has an interest in video... Read More →
avatar for Richard Hanley

Richard Hanley

Senior Software Engineer, Google
Richard is a software engineer at Google, and a newcomer to OCP. He primarily works on Open BMC, and is interested building a richer model of computer systems to help develop/test the control plane.He has a background in consumer electronics, both as an embedded engineer and as a... Read More →


Wednesday May 13, 2020 11:30am - 11:45am
EW: Systems & HW Management

11:30am

An Advanced Liquid Cooling Design for JD.COM Data Center
As the emerging business models and advanced customer experience driving an explosion of data at the era of 5G, data center has been evolving with increasingly powerful AI (Artificial Intelligence) computing capability for enhancing CSP (cloud service provider) cloud service by data analytics in data centric world. High-performance high-power computing, storage and networking are backend behind this. Meanwhile, having data center greener for energy efficiency is mandatory design target for modern data center, measured with PUE. Advanced thermal solution is one of crucial technology to continually address the demand of higher performance computing at lower PUE.

D2C (Direct to Chip) liquid cooling, as a typical advanced thermal solution, has been accepted by the industry as an effective cooling solution for high density datacenters. JD.com has years’of liquid cooling engineering design experience from rack level to data center level. This presentation introduces best known method (BKM) from JD.com engineering practice and deployment data. It also focuses on system level liquid cooling design related ingredients such as CPU cold plates, memory cold plates, leakage detection, quick disconnects, etc. Moreover, this presentation details a JD.com own advanced design methodology by engineering practice data analytics, optimal for CPU cold plate design. Also, whole-loop leakage detection has been considered as one of the pain points in the industry, this presentation will also discuss the implementation of whole-loop leakage detection in a JD datacenter.

Speakers
NA

Nishi Ahuja

Senior Principal Engineer, Intel


Wednesday May 13, 2020 11:30am - 12:00pm
EW: Advanced Cooling Solutions

11:30am

Re-imagining SSD Storage with Cloud-Scale Innovation - presented by Samsung
Today’s datacenters and data lake sizes call for re-imagining traditional storage systems to increase efficiency and scalability. New, data-centric approaches avoid redundant address translations, minimize protocol terminations, and act as intelligent storage subsystems. As an industry leader in solid state storage, we’ll spotlight the latest advancements in SSDs that will dramatically impact Cloud deployments in 2020 and beyond. Some examples include discussion of Ethernet-attached SSDs, computational storage, Zoned Namespaces, and key-value indexed SSDs.

Speakers
avatar for Michael Smullen

Michael Smullen

Sr. Manager Datacenter SSD Products Group, Samsung
Cradle to grave product owner: definition, MRD/PRD, development, rollout, early customer programs, pricing strategy and tactical negotiation, design win budget achievement, revenue and gross margin management, and EOL analysis and execution.Specialize in technology products in the... Read More →
avatar for Pankaj Mehra

Pankaj Mehra

VP Storage Pathfinding, Samsung
Pankaj Mehra is VP of Product Planning at Samsung Electronics. He was previously VP and Senior Fellow at both SanDisk and WD, WW CTO at Fusion-io, and a Distinguished Technologist at HP. Besides his 3 books on Machine Learning and Systems, Pankaj has over 50 papers and patents resulting... Read More →


Wednesday May 13, 2020 11:30am - 12:00pm
Executive Tracks

11:30am

Reducing the Cost of Content Distribution in Africa
In many areas of Technology, African businesses have jumped the curve, partly due to not having invested in stepping-stone technologies. However, in the Data Centre, Vanity Brands have held a stronghold that has held back content development and distribution at a price that is consumable by over a billion Africans.

Using its own Case Studies based on implementation of Open Compute infrastructure, Atlancis is working with Mobile Network Operators, Banks and Service Providers to reduce the time and cost of distributing valuable content, and in so doing, democratising the internet for Africa.

This presentation will take the audience on Atlancis' OCP journey and look at the future role OCP will play in developing Africa's technology landscape.

Speakers
avatar for Paul Statham

Paul Statham

Regional Commercial Director, Atlancis
Speaker, Investor, ICT Executive passionately opening Africa's eyes to the benefits of OCP.


Wednesday May 13, 2020 11:30am - 12:00pm
OCP Case Studies

11:40am

Signal Integrity Challenges for PCIe 5.0 OCP Topologies
This presentation explores the signal integrity challenges of PCIe 5.0 and the corresponding OCP system design challenges, where the right balance must be found between PCB materials, connector types, and the use of signal conditioning devices for practical compute topologies: CPU-to-AIC with one/two connectors, JBOG accelerator module baseboard, etc. Each topology has multiple factors impacting SI—package, channel insertion/reflection loss, and crosstalk—requiring tradeoffs between physical channel length, interconnect solution, PCB design/material, etc. Through an objective analysis using open-source tools, the goal is to provide the audience with a step-by-step methodology to optimize their topology between low-loss PCB material, alternative connectors, signal retimers, or some combination thereof. Finally, to improve link uptime and maximize the potential of PCIe 5.0 throughput and latency, we explore the relationship between Lane BER and Link stability.

Speakers
avatar for Casey Morrison

Casey Morrison

Head of Systems and Applications, Astera Labs
Casey Morrison leads Applications Engineering at Astera Labs and is responsible for product definition and customer design-in activities. With 12+ years of experience in high-speed interfaces for data center and wired/wireless communications systems, he has a passion for creating... Read More →
avatar for Pulkit Khandelwal

Pulkit Khandelwal

Technical Staff Member, Astera Labs
Pulkit Khandelwal (pulkit.khandelwal@asteralabs.com) leads Systems Engineering at Astera Labs, working on system applications, software/firmware development, and system validation. He has over 15 years of experience in data-center connectivity solutions, developing and deploying leading-edge PHYs and ASSPs for electrical and optical inte... Read More →


Wednesday May 13, 2020 11:40am - 11:55am
EW: Servers

11:50am

Firmware Security Without Obscurity
Firmware has been growing in size and complexity over the years and has become a rich target for attackers. Firmware is currently a popular topic in security research and we find that our work in OSF overlaps some of the work being done in the OCP Security Working Group.

With data rapidly being added or moved to the cloud, hyperscalers and SMBs alike are facing a massive security challenge and often rely on firmware and tools they have little or no visibility into. We assert that security should not be obscure; it should be open, transparent, and available to all.

Furthermore, the hyperscaler model fundamentally changes the way we view firmware development and the supply chain. As hyperscalers continue to grow it has become increasingly important to ensure control and accountability throughout the server lifecycle starting with design and throughout development, validation, deployment, and eventual decommissioning.

In this talk we will discuss design principles, available features, and will walk through a real example of an open and transparent security solution running with OSF on OCP hardware to show how one can build, measure, and verify the integrity of the firmware running on their system. We will also discuss areas where OCP and industry can help white hat research and development efforts.

Speakers
avatar for David Hendricks

David Hendricks

Firmware Engineer, Facebook
David got his first taste of BIOS tweaking as a kid overclocking gaming PCs and was drawn into the coreboot world where he learned real firmware hacking. Since then he has developed, deployed, and advocated for open source firmware in consumer electronics and datacenter infrastructure... Read More →
avatar for Philipp Deppenwiese

Philipp Deppenwiese

Head of Cyber Security and Firmware Department, 9elements Cyber Security
CEO of the immune GmbH in Germany. Head of the Cyber Security and Firmware Department of 9elements Agency GmbH in Germany. Founder of the Open Source Firmware Conference. coreboot and LinuxBoot enthusiast.


Wednesday May 13, 2020 11:50am - 12:10pm
EW: Open System Firmware

11:55am

Isolated Control Plane for OpenRack v3 Servers
We introduce a new architecture designed for secure multi-tenant cloud native computing workloads. Alabama is a stateless Open Rack v3 server featuring an isolated control plane separating the operator and tenant. The operator manages the server and all the I/O cards (e.g., Smart NICs and local storage controllers) via a platform controller, allowing untrusted tenants to run on the host CPUs only using resources configured by the operator. The platform controller manages all the I/O cards (Smart NICs, local storage controllers) with a standardized management API. The I/O cards are connected to the platform controller with a new standardized internal connector on the PCIe CEM form factor carrying either PCIe or Gigabit Ethernet lanes.

The isolated control plane provides secure, high-performance management at scale. Open management APIs and open hardware interface definitions will enable a broad ecosystem of I/O cards.

Speakers
avatar for Alex Fuxa

Alex Fuxa

Section Manager - Cloud Native HW & Smart NICs, HPE
Alex Fuxa is a section manager in the Advanced Development Group within the Compute Business Unit. Alex\'s team focuses on two area\'s of technology: Cloud Native Hardware and Smart NICs. In this role he contributes by setting technical direction, contributing as an architect, and... Read More →
avatar for Robert Elliott

Robert Elliott

Technologist, HPE
Rob has worked on ASIC design and verification, firmware, drivers, and system architecture for Compaq/HP/Hewlett Packard Enterprise laptops, workstations, servers, and storage products. He created the Serial Attached SCSI (SAS) standard and helped introduce persistent memory to the... Read More →
avatar for Scott Shaffer

Scott Shaffer

VP, Advanced Development, HPE


Wednesday May 13, 2020 11:55am - 12:10pm
EW: Servers

12:00pm

Eco-System Enabling of Liquid Cooling Ingredients
The continued demand for increased compute performance comes with an increase in server power and power density. The increased server power requires more efficient cooling capabilities compared to traditionally used air cooling. Direct liquid cooling using cold plates can provide the needed cooling capability, as well as fit within a small volume to allow for greater server and compute density in the rack. Even though liquid cooling is not a new cooling strategy for IT equipment, the industry is not ready to efficiently mass deploy liquid cooling solutions. Therefore, Intel is working to enable liquid cooling ingredients with industry partners. The intent is to create more available, lower risk, and faster time to market opportunity for OEMs to deploy liquid cooling solutions. Intel’s efforts are specifically focused on manifold distributed liquid cooling at the server rack level (i.e. the technology cooling system, TCS). The TCS is the fluid loop providing the cooling liquid from the Coolant Distribution Unit (CDU) through the rack manifold to the IT equipment cold plates, and then back again to the CDU through the rack manifold. The liquid cooling ingredients in the secondary loop included in Intel’s enabling efforts are: cooling liquid, CDU, interchangeable Universal Quick Disconnects (UQD), and cold plates. Intel’s enabling efforts, together with Intel’s strategy to enable these ingredients was first introduced at the OCP Global Summit 2019, and an update on the eco-system enabling status will be presented at the OCP Global Summit 2020.

Speakers
avatar for Jessica Gullbrand

Jessica Gullbrand

Tech Lead Thermal Engineer, Intel
Jessica is a Thermal Tech Lead Engineer at Intel in the Data Platform Group (DPG). She is currently working on energy efficiency, immersion, and liquid cooling cold plate technologies. Jessica has worked on new and innovative technology solutions in fluids, acoustics, and thermals... Read More →
avatar for Mark Sprenger

Mark Sprenger

Tech Lead Engineer, Intel
Mark Sprenger is Senior Mechanical Engineer and Technical Lead in Data Platform Group responsible for external development of cooling solutions for commercial datacenter products. Mark has over 25 years of industry experience in advanced development covering a wide variety of technologies... Read More →


Wednesday May 13, 2020 12:00pm - 12:30pm
EW: Advanced Cooling Solutions

12:00pm

Bringing the F16 Network into the Lab
The new Facebook DC Topology, F16, introduced at the 2019 OCP Global Summit. Hardware, protocols, and distributed systems, require evaluation, development, and validation before deployment. This is a significant effort when done in a Facebook Data Center, and happens late in the cycle. Facebook Network Hardware and the Broadcom Switch organization jointly created an alternative: collapsing the F16 network into a single switch. A DC Network can be conveniently recreated in the lab at Facebook, at vendor’s facilities, and at ODMs. We called it F16InABox.

F16InABox recreates all F16 stages (RSW, FSW, SSW) inside a 12.8Tbps Broadcom Tomahawk 3 switch with ECMP groups, oversubscription ratios, and hashing logic between these stages matching F16. The system has about 4.8T for server endpoints (about 2 racks equivalent) where the logical placement of a server can be rack local, POD local, or cross-POD with respect to any other server. The multi-stage/multi-path DC network queuing behaviors appear in F16inabox without having to use a control plane like BGP.

F16InABox can self-generate background traffic without requiring external traffic generators, and uses timestamping to accurately measure end-to-end delay for every packet.

We used F16inabox to evaluate very fast distributed systems consensus and consistency primitives, and will discuss other uses like
- PTP performance evaluation
- Storage over the network
- NIC feature evaluation and validation
- ODM integration testing
- Latency reduction mechanisms (AQM)

Speakers
AB

Ahmad Byagowi

Network Hardware Engineer, Facebook
PN

Paul Ng

Network Hardware Test Automation Development Engineer, Facebook
avatar for Surendra Anubolu

Surendra Anubolu

Distinguished Engineer, Broadcom
Ethernet Storage HPC AI networking


Wednesday May 13, 2020 12:00pm - 12:30pm
EW: Networking Hardware

12:00pm

Evolving Adaptive Platform for High Performance Computing - presented by QCT
- QCT Company profile
- Industrial HPC Workloads and optimization
- Platform on Demand - QCT POD building blocks design and containerized
- Case study on QPOD for HPC - HPC workloads

Speakers
avatar for Stephen Wheat

Stephen Wheat

Senior Director/Consultant - Business Group, QCT
35+ Years of experience using, designing, and developing High Performance Computing systems for both, large and small facilities. 20 years of that time was at Intel, where I was instrumental in rebuilding the HPC business efforts, becoming General Manger of that business. I expect... Read More →


Wednesday May 13, 2020 12:00pm - 12:30pm
Executive Tracks

12:00pm

Case Study: Water-Cooling in OCP Servers, Racks & Data Centers
Cloud&Heat Technologies first implemented OCP rack and server design preliminary specifications in a customer project in 2019. The result was a 40 ft mobile data center container with a power density of 400 kW and equipped with 125 innovative, in-house developed, water cooled server chassis for server hosting and rendering farm purposes. The compact rack construction was adapted in accordance with the OCP 21ʺ standard. The backplane was designed to provide an interface with the power supply as well as a connection to the cooling system, enabling flexible and snap-in insertion of the chassis. With a design PUE of 1.08 the container saves more than 300,000 kWh electric energy and over 200 t CO2 per year compared to a conventional datacenter. With German energy prices that leads to a saving of more than 60,000 $ per year. This keynote will give an insight into the design challenges met and practical experience gained during planning, installation and operation of the prototype. Focusing on potential contributions to the OCP project, I will present Cloud&Heatís latest developments in the field of water cooling. I will also show direct water-cooled power supply units with a 3 kW and 15 kW power supply. All in all, Cloud&Heat has 4,500 servers now in the field. Our company is focused on sustainable and green IT hardware and software services; having evaluated a variety of cooling solutions, we found hot-liquid cooling to be the best and most efficient way of realizing heat recovery and seamless building integration.

Speakers
avatar for Jens Struckmeier

Jens Struckmeier

CTO and Founder, Cloud&Heat Technologies
Dr. Jens Struckmeier is a physicist, inventor and entrepreneur from Germany. Whilst still a student, he was responsible for the planning of the first passive house in North Rhine-Westphalia, integrating an innovative heating system. Between 2000 and 2003 Jens worked as a project lead... Read More →


Wednesday May 13, 2020 12:00pm - 12:30pm
OCP Case Studies

12:00pm

LOG IN HERE for the Virtual Summit!
LOG IN HERE: https://onlinexperiences.com/scripts/Server.nxp?LASCmd=L:0&AI=1&InitialDisplay=1&ClientBrowser=0&ShowKey=86206

REGISTER HERE: https://www.opencompute.org/summit/global-summit/registration

Wednesday May 13, 2020 12:00pm - 4:00pm

12:10pm

Next-Gen Accelerator, NIC, and Compute Using CXL
In this workshop, we will focus on device types and capabilities of devices supporting Compute Express Link (CXL), state of the ecosystem support around CXL, and how CXL capabilities can be leveraged for future OCP projects in the areas of compute, accelerator, NIC, and memory expansion. The call to action is for OCP sub projects in server (EG: NIC, OAI/Open Accelerator Infrastructure) to incorporate support for CXL in future products and platforms.



Speakers
avatar for Glenn Ward

Glenn Ward

Senior Director of Technology Development, Microsoft Azure; CXL Consortium™ Marketing Work Group Co-Chair, Microsoft
Glenn Ward is a Senior Director of Technology Development for Microsoft Azure, where he helps guide development of accelerators and platform capabilities to improve Azure services. Glenn drove the development of Microsoft’s first Cloud ASIC, designed to radically lower Cloud storage... Read More →


Wednesday May 13, 2020 12:10pm - 12:25pm
EW: Servers

12:10pm

Minimum Platform: Open Source Intel UEFI Firmware
Firmware for the majority of the Intel product portfolio is implemented using Unified Extensible Firmware Interface (UEFI) compliant firmware. Historically, UEFI firmware was built with a PC ecosystem mindset; open standards but closed implementations.

The emergence of the cloud has shifted the development of computing systems from large-scale OEMs to individual end users. Accordingly, open standards by themselves are no longer enough, modern implementations must be both open source and standards compliant. The key benefit of firmware standards like UEFI is broad operating system and peripheral compatibility, and the ability to mass deploy a single OS image on virtual machines or bare metal machines without modifying the OS image or the firmware.

This presentation outlines the TianoCore Minimum Platform Architecture (MPA), which Intel designed to facilitate a simplified, modular, and consistent firmware porting process. This makes it easier to produce high quality, open source firmware optimized for cloud systems. The MPA Staged Boot Approach will be introduced, which breaks firmware functionality into two main categories:

1. Minimum components required for authenticated boot.

2. Advanced features as modular components, individually selected and customized for the end user's system.

This session will also cover advanced features that are available today including IPMI, USB 3.0 Debug, SMBIOS, and Networking. The presentation will close with a call for collaboration and information on how to get started.

Speakers
ND

Nate DeSimone

Firmware Engineer, Intel
Nate DeSimone is a firmware engineer at Intel with broad experience in firmware development for several generations of embedded, mobile, client, and server products. Nate composed the majority of the Intel FSP 2.1 Specification, is an active contributor to TianoCore, and is an advocate... Read More →


Wednesday May 13, 2020 12:10pm - 12:30pm
EW: Open System Firmware

12:25pm

Accelerating Machine Learning in OCP Platforms
Advancing machine learning (ML) architectures generally involves the partnership of hardware designers, data scientists, and optimization engineers. In this talk, we review trends in numerical precisions used in ML architectures. We detail advances in the Intel Xeon processor used across various OCP platforms that accelerate ML workloads. Specifically we present new instructions that accelerate 16-bit and 8-bit operations. We share an update on projects where Intel is collaborating with the ML OSS community to accelerate AI performance across various OCP platforms.

Speakers
avatar for Andres Rodriguez

Andres Rodriguez

Sr. Principal Engineer, Intel
Andres Rodriguez is a Sr. Principal Engineer in the Data Center Group (DCG) where he designs deep learning solutions for Intel’s customers and provides technical leadership across Intel for deep learning products. He has 15 years of experience working in artificial intelligence... Read More →


Wednesday May 13, 2020 12:25pm - 12:45pm
EW: Servers

12:30pm

Xeon SP FSP and Coreboot Status Update
This joint presentation between Intel and Facebook gives an update on FSP and coreboot for Xeon SP (specifically Skylake Scalable Processor and Copperlake Scalable Processor) .

Speakers
avatar for Anjaneya (Reddy) Chagam

Anjaneya (Reddy) Chagam

Senior Principal Engineer, Intel
Anjaneya “Reddy” Chagam is a Senior Principal Engineer and Chief Software-Defined Storage Architect in Intel’s Data Center Group. He is responsible for developing software-defined storage strategy, architecture, and technology initiatives. He is a member of the Technical Steering... Read More →
avatar for Jonathan Zhang

Jonathan Zhang

Software Engineer, Facebook
Jonathan is passionate at architecting lower level system firmware/software to achieve computer system design goals. He used to work at software organization of hardware companies (silicon vendors), he is working at hardware organization of software company (hyper-scaler). He is thrilled... Read More →


Wednesday May 13, 2020 12:30pm - 12:50pm
EW: Open System Firmware

12:30pm

Optimized immersion for OCP Server Systems
Workshop about the latest developments from Asperitas, the adoption of a new form factor, the solutions which have been integrated in the Asperitas technology and the compliancy with OCP Immersion Requirements.

Speakers
avatar for Rolf Brink

Rolf Brink

CEO, Asperitas
Product development for cloud architectures and datacentre infrastructure, combined with a passion for cleantech innovations like liquid cooling, are the foundation of the developments which drive Asperitas.Asperitas is a clean-tech company focused on greening the datacentre industry... Read More →


Wednesday May 13, 2020 12:30pm - 1:00pm
EW: Advanced Cooling Solutions

12:30pm

Enabling Secure Boot in ONIE
Platform resiliency is a necessity in systems that are manufactured today, be it for networking devices, servers or IOT devices. Systems should be built in a way that allows the authenticity of the entire software stack to be verified. This is only possible by starting with a Root of Trust that is immutable and thereafter each subsequent stage verified by the previous stage. All possible boot options on the system should be authenticated. ONIE is used to install Network Operating Systems and is inherently present on most networking devices that use ONIE for installation. This presentation will show how ONIE secure boot can be enabled while also showing all the other secure boot components in the chain of trust. What already exists in open source ONIE is enabling Shim for verification of Grub and Kernel. However, Shim cannot be used to verify grub.cfg or initial ramdisk. Grub however has the option to verify Linux Kernel, initial ramdisk and grub.cfg. Using this concept, BIOS is used to verify Grub and Grub to verify the next stage it loads. Grub installation in ONIE installer scripts cannot use existing Grub installation methods because of the necessity to use signed standalone Grub. Also, Key enrollment scripts in ONIE will be used to add or revoke keys in UEFI key database. It can be run before calling NOS installer or ONIE updater. This concept of enabling Secure Boot in ONIE can also be used as an example to enable Secure Boot in any NOS.


Speakers
avatar for Hannah Williams

Hannah Williams

Software Senior Principal Engineer, Dell Technologies
Hannah is currently working on Dell Networking platforms with primary focus in BIOS. The last few years she worked on enabling both UEFI BIOS and Coreboot for platforms running on Intel SOCs.https://www.linkedin.com/in/hannah-williams-219849/


Wednesday May 13, 2020 12:30pm - 1:00pm
EW: Networking Hardware

12:30pm

Unlocking the Power of Circular Data Centers - presented by ITRenew
In response to the unprecedented scale of current and forecasted infrastructure demands, the largest data center owners in the world have leveraged open hardware to drive major advances in TCO and optimize their corresponding hardware refresh cycles. But challenges remain. Big challenges. First, this same opportunity has not been available to the broader global IT markets; and second, despite this progress, exponential growth in data and compute requirements are driving infrastructure scale that is unsustainable.A circular economic IT model is the answer to both. In this session, Ali Fenn, President of ITRenew, will share real-world data, customer experiences and products that maximize the financial and sustainability returns of your IT hardware today. Whether you already run massive data centers, or are headed that way and facing capex and resource constraints as you scale, adopting circular principles and solutions can significantly decrease IT hardware TCO, and minimize your CO2e impact. Together, let’s use our tremendous collective power to democratize access to OCP solutions for enterprises globally, maximize the lifetime value of assets, and achieve an essential sustainability advantage.

Speakers
avatar for Ali Fenn

Ali Fenn

President, ITRenew
Ali Fenn is the President at ITRenew, where she oversees all revenue and leads the Company’s circular data center initiatives, including market development and business model innovation. Open hardware platforms and open source software innovation are the critical foundation underpinning... Read More →


Wednesday May 13, 2020 12:30pm - 1:00pm
Executive Tracks

12:30pm

Modernizing Campus Networking
This presentation will talk about emerging trends in campus networks and the benefits of a common operational model for data center and campus networks. It will also address automated network validation and troubleshooting techniques for campus networks and the use of modern CI methods for automated deployment of new software and configurations.

Speakers
avatar for Partho Mishra

Partho Mishra

President and Chief Product Manager, Cumulus Networks
Partho Mishra is President & Chief Product Officer at Cumulus Networks. Prior to joining Cumulus, Partho was running the RASA Network Analytics team at Aruba/HPE, following the acquisition of RASA by Aruba in 2016. Previously, he was Vice President and General Manager of Cisco’s... Read More →


Wednesday May 13, 2020 12:30pm - 1:00pm
OCP Case Studies

12:45pm

Designing Balanced System for Small to Large Scale Deeplearning Training
Deep learning training workloads call for a balanced system design with an optimal balance of compute, acceleration and memory.  In this talk we discuss how we leverage system design knobs across compute, memory, storage & networking, coupled with the systems & hardware management tools and software to develop an optimized solution for customers training needs from small scale workloads running on a few cards to large scale complex workloads spanning multiple chassis & racks. We will showcase how the flexibility of using different fabric solutions along with the Inter Chip Link enables different topologies and highly efficient training system configurations to be built. We will share how we design to strike a balance between system performance, software complexity and total cost of ownership.

Speakers
avatar for A.K. Roy

A.K. Roy

Director, AI Training Systems & Solutions Strategy, Training Products Group (TPG), AIPG, Intel
A.K. Roy is the Director of AI Training Systems & Solutions Strategy & Product Management, in the Training Products Group (TPG) at Intel Corporation’s AI Platforms Group (AIPG). He has over 15 years of experience in AI, Cloud and Server Systems & Solutions Strategy, Product management... Read More →


Wednesday May 13, 2020 12:45pm - 1:00pm
EW: Servers

12:50pm

Enable OpenTitan on FPGA with Oreboot
In this talk we will demonstrate how to enable OpenTitan, an open source silicon Root of Trust (RoT) project (which originally is an dedicated security chip designed for servers), to be enabled in FPGA. This presentation We will also describe how latest innovative boot solution – Oreboot could provides the best infrastructure to support OpenTitan project. Oreboot is a fully open-source power-on-reset and romstage firmware written in Rust, and requires all support packages to be open-sourced by design. By using the Rust programming language, Oreboot has a leg-up in terms of security and reliability compared to contemporary firmware written in C or assembly. Hence OpenTitan and Oreboot are perfectly fitted together as a comprehensive solution to address the OCP standards and industry requirements.

Speakers
avatar for Ron Minnich

Ron Minnich

Software Engineer, Google
linuxboot, u-root, coreboot, linuxbios, ... all open source firmwarelinux kernel, servers,
avatar for Andi Azfar

Andi Azfar

Engineer, Intel
Andi is a System Validation engineer working on Intel FPGAs after joining Intel in 2018. Andi comes from a different engineering background but is currently interested in all things FPGA and Data Science. Outside of work Andi likes to play sports but especially enjoys playing Frisbee... Read More →
avatar for Tan Lean Sheng

Tan Lean Sheng

Software Engineer, Intel
Sheng is a firmware engineer working on coreboot and Slim Bootloader projects after joining Intel in 2017. Since then he has developed a keen interest in firmware development and actively involved in open source firmware initiatives . Outside of work Sheng is a tech geek and enjoys... Read More →


Wednesday May 13, 2020 12:50pm - 1:20pm
EW: Open System Firmware

1:00pm

Small JBOG Solutions
There are 2 JBOG solutions now with OCP (Big Surr and Big Basin) and they are heavy. The goal of my project is to bring a small JBOG solution to the OCP community.

This JBOG would be working with a head node such as a Leopard or Tiago Pass servers. Up to 4 x16 GPU could be connected to it (so 4 GPU per servers). This JBOG would be compatible with both Open Rack V1, Open Rack V2 and the Discovery chassis (project that ITRenew presented last year)

I would like to present the steps of my project, the design choices, people/company who might be interested and also get some feedback from the community to know what could be changed or improved on the JBOG

Speakers
avatar for Thomas Van Liefde

Thomas Van Liefde

Electrical Engineer, ITRenew


Wednesday May 13, 2020 1:00pm - 1:15pm
EW: Servers

1:00pm

Recent Developments in Immersion Cooling with Fluorochemical Fluids
This presentation will follow the evolution of the technology by highlighting several key immersion deployments from the past 10 years. A cost of ownership analysis suggests value for hyperscale data centers. It concludes with a summary of engineering and supply chain challenges on the path to more widespread adoption.

Speakers
PT

Phillip Tuma

Application Development Specialist, 3M


Wednesday May 13, 2020 1:00pm - 1:30pm
EW: Advanced Cooling Solutions

1:00pm

Advanced Ethernet Telemetry Working Across NOS’s and Switches
Squeezing the most performance from modern Cloud Datacenters and Machine Learning pods requires Ethernet NICS and switches that provide advanced levels of visibility. Mellanox WJH leverages hardware-accelerated network telemetry functions that work across multiple Network Operating Systems, including SONiC, Cumulus, Onyx, and Native Linux. Advanced telemetry requires a presentation layer that does more than just sift & filter alerts, so this solution works with open source tools like Grafana and OpenStack, as well as third party orchestration solutions from Apstra, Cumulus Networks, and Xcloud Networks. Join us as we discuss workload-speeding telemetry and see a demo of 4 different NOS’s sending WJH notifications that get graphically displayed across a variety of tools that leverage WJH for datacenter-wide visibility.

Speakers
avatar for Alex Saroyan

Alex Saroyan

CEO, Netris
CEO of Netris - the new automatic netops platform that turns your physical network into service, and provides network services to your Kubernetes cluster, compute, and storage.
avatar for Justin Betz

Justin Betz

Technical Marketing Engineer, Cumulus Networks
Ask me about Cumulus NetQ and network operations with CI/CD!
avatar for David Iles

David Iles

Sr. Director, Ethernet Switch, Mellanox
David Iles, Senior Director of Ethernet Switching at Mellanox Technologies:Over the last 20 years, David has held leadership roles at 3COM, Cisco Systems, Nortel Networks, and IBM where he pioneered advanced network technologies including High Speed Ethernet, Layer 4-7 switching... Read More →


Wednesday May 13, 2020 1:00pm - 1:30pm
EW: Networking Hardware

1:00pm

From OCP Hardware to OCP Total Solution - presented by Inspur
Inspur is striving for a better value-add solution to the public cloud & hybrid cloud customers. The total solution OCP product is one of the key elecments to make this happen.  In this session, we will talk about POCs, refer cases on SAP HANA, and in-memory DB application running on OCP product Crane Mountain.  Then the session will dive into the benefits, from the performance per unit height and performance per watt.

Speakers
avatar for Yinan Liu

Yinan Liu

Customer Program Manager, Inspur
Yinan is a Customer Program Manager for Inspur’s strategic CSP customers. He is leading the product development of Inspur’s open hardware platforms and actively driving the solution innovation with partners.


Wednesday May 13, 2020 1:00pm - 1:30pm
Executive Tracks

1:00pm

Expanding OCP Gears at Yahoo! JAPAN
Since 2016, we have adopted more than 4,000 OCP servers in our data centers. We will explain how we approach OCP deals for adoption and how we integrate and operate OCP servers and other related components in our infrastructure. Specifically, we will explain the day to day activities of YJ's server team and some of the OCP initiatives we are working on. We will also explain the differences between Japan and other countries related to the adoption of OCP. We will use a series of metrics to compare OCP servers and EIA19(Standard Servers), including CAPEX and OPEX, to highlight the benefits of OCP adoption. Since adopting OCP servers our the data center has changed more than ever. We will talk about why it has changed, what has changed and how this has improved the mindset and facilities at yahoo japan.

Speakers
avatar for Kazuhide Fujimi

Kazuhide Fujimi

Server Infrastructure Architect, Yahoo Japan Corporation
Kazuhide Fujimi is a Server Infrastructure Architect at Yahoo! Japan and has responsibility for procurement, testing, QA and troubleshooting of server hardware. Yahoo! Japan is one of the famous CSP in Japan and has several data centers in Japan as on-premise Infrastructure envir... Read More →
avatar for Keita Uchida

Keita Uchida

Senior Server Engineer, Yahoo Japan Corporation
avatar for Cheng Zhao

Cheng Zhao

Senior Datacenter Engineer, Yahoo Japan Corporation


Wednesday May 13, 2020 1:00pm - 1:30pm
OCP Case Studies

1:20pm

Introducing Intel® Server PCH Ignition Firmware
Intel recently has provided Server PCH ME firmware binary under a redistributable license to the edk2-non-osi repository. As some immediately noticed the size of this firmware is significantly smaller (0.5MB) comparing to original Intel® Server Platform Services (SPS) ME firmware size that takes around 3MB on SPI flash. How this was possible and how the functional decomposition was made to support the leaner version of Server PCH ME firmware? How does this align with Open platforms and Intel’s efforts for Open Firmware? This transition triggered detailed analysis that resulted in functional split between multiple server platform components. As a result, the Node Manager power control real-time function was functionally decomposed between various platform components. Same approach was applied to other firmware functions previously available in Server PCH ME firmware like telemetry, PCIe MCTP infrastructure support. As a result Server PCH Ignition firmware is responsible only for HW platform initialization and support for Intel® Bootguard technology. This approach enables us to reduce the size of SPI partition designated for SPS image by 6 times so full redundant PCH Ignition SPI sub-partition fits into 0.5MB.

Speakers
avatar for Mariusz Oriol

Mariusz Oriol

Principal Engineer, Intel
avatar for Piotr Sawicki

Piotr Sawicki

Firmware Architect, Intel
Piotr Sawicki is a Firmware Architect at Intel Corporation Data center Engineering Group in Poland where he has been working for more than a decade on development of server FW solutions. He was an architect of Server Platform Services (SPS) Node Manager and Automotive ME FW. Piotr... Read More →


Wednesday May 13, 2020 1:20pm - 1:35pm
EW: Open System Firmware

1:30pm

Design Guidelines for Immersion-Cooled IT Equipment
The presentation of the IT gear for Immersion whitepaper which is scheduled for completion during the summit.

Speakers
avatar for Rolf Brink

Rolf Brink

CEO, Asperitas
Product development for cloud architectures and datacentre infrastructure, combined with a passion for cleantech innovations like liquid cooling, are the foundation of the developments which drive Asperitas.Asperitas is a clean-tech company focused on greening the datacentre industry... Read More →
avatar for Jessica Gullbrand

Jessica Gullbrand

Tech Lead Thermal Engineer, Intel
Jessica is a Thermal Tech Lead Engineer at Intel in the Data Platform Group (DPG). She is currently working on energy efficiency, immersion, and liquid cooling cold plate technologies. Jessica has worked on new and innovative technology solutions in fluids, acoustics, and thermals... Read More →
avatar for Jimil Shah

Jimil Shah

Application Development Engineer, 3M
Immersion Cooling of Data Centers
RP

Rick Payne

Sr. Director, Systems Engineer, Flex
avatar for Leonard CIUBOTARU

Leonard CIUBOTARU

Special Projects Division Manager (R&D), 2CRSi
Leonard Ciubotaru is the Special Projects Division manager, within the R&D Department of 2CRSi, where he oversees all long term, innovative research projects together with unique on-demand customer requests. He is also the technical adviser for OCP (mechanical) related projects. He... Read More →


Wednesday May 13, 2020 1:30pm - 2:00pm
EW: Advanced Cooling Solutions

1:30pm

On-Board Optics Blade for Minipack Fabric Switch
A 1.6 Tb/s blade for Facebook’s Minipack fabric switch is designed based on four, 400Gbps on-board optics modules developed by the Luxtera team at Cisco, with each module supporting four 100G CWDM4 ports. This on-board optics blade is a direct replacement for the standard 1.6 Tb/s Minipack blade based on 16x100G QSFP optical transceivers. This represents the first deployment of on-board optics in Facebook data centers.

Designing a blade with on-board optics promises several benefits over traditional pluggable optics. As the form factor may be adapted for the intended application, a less constrained design can be optimized for the system as a whole. Trade-offs between number of ports integrated within a module and packaging cost per port can be weighed against manufacturability. Additionally, on-board optics provide more flexibility in the design of the high-speed electrical link between the switch ASIC and the optics. Whereas pluggable optics require the electrical path to be routed to the front panel of the switch box, on-board optics gives the flexibility to place the optics on the main board closer to the switch ASIC, allowing for power savings through the elimination of retimers and the ability to specify lower power SERDES on the electrical links.

While pluggable optics have been used extensively in hyper-scale data centers, on-board optics present new operational challenges. This work is also motivated by the desire to develop the tools needed to operationalize on-board optics in our data center environment.

Speakers
CW

Christopher W. Berry

Optical Engineer, Facebook


Wednesday May 13, 2020 1:30pm - 2:00pm
EW: Networking Hardware

1:30pm

Transition to OCP Infrastructure in a Telco Environment
As a Hong Kong based carrier that has built orchestration software to help other Telcos convert their legacy business to Overlay based SDN network, DCConnect has highly demanding networking requirements that necessitated a move from legacy, vendor centric switching to an OCP-based white-box model.  In this presentation, Jonathan will share our experience of our transition to that model, including a brief overview of the company, the requirements that sparked the change, the original state of our network, the transition to our OCP-based current state, and where we will go from there.


Speakers
avatar for Jonathan Leung

Jonathan Leung

Head of R&D, DCConnect
Jonathan is the Head of R&D at DCConnect Global, which is a pioneer in innovating on-demand global cloud connectivity.  With a passion for NetOps and DevOps he leads a team of talented developers that has developed DCConnect’s orchestration platform, multi-vendor control plane... Read More →


Wednesday May 13, 2020 1:30pm - 2:00pm
OCP Case Studies

1:30pm

OCP Impact Study 2020: The Emerging Edge, Circular Economy Embers Glow
At the close of 2019, Omdia (formerly IHS Markit) conducted a survey of top OCP vendors and end users to determine growth of 2019 non-board member adoption over 2018, an update to their 2017 study. In addition, they asked respondents to share how the market has changed over the last 12 months covering technology demands, shifting compute to the edge, circular economy, distribution models, and more. Please join this session to learn the details of their findings where they will discus how the market has matured and shifted from last year, take a look at 2019 results vs last year’s and provide and updated forecast to 2023.

Speakers
avatar for Vladimir Galabov

Vladimir Galabov

Principal Analyst, Omdia (formerly IHS Markit)
As a principal analyst in the cloud and data center research practice, Vlad focuses on data center compute covering disruptive trends such as the shift to cloud computing, innovations in the Ethernet adapter and server markets, the adoption of Open Compute Project (OCP) equipment... Read More →
avatar for Cliff Grossner

Cliff Grossner

Executive Director Research & Technology Fellow, Omdia (formerly IHS Markit)
\"As head of the Cloud and Data Center Research Practice, he is responsible for research quality across the practice and his research is focused on cloud services, data center compute and networking, and software defined networking (SDN). Dr. Grossner has more than 25 years of telecommunications... Read More →


Wednesday May 13, 2020 1:30pm - 2:15pm
Executive Tracks

1:35pm

Enabling Slimbootloader with LinuxBoot on Intel MicroServer Platform
While server industry is rapidly changing, it has become a real challenge for us to keep the pace to provide full-featured and high-quality system firmware solution to quickly meet customer’s needs. On the other side Linux kernel has a huge development community and it covers many duplicated functions with system firmware. So why not utilize the standard kernel capability to do things that was done by system firmware? LinuxBoot is an open sourced solution targeted for this purpose, and it is gaining momentum among industry players and open source community as they are seeking for the next generation of firmware technology. This presentation describes how Intel’s latest innovative boot solution – Slim Bootloader provides the best infrastructure to support LinuxBoot so as to simplify the boot firmware, and how it fits the OCP standards and industry requirements.

Speakers
DG

David Guckian

Software Engineer, Intel
avatar for Tan Lean Sheng

Tan Lean Sheng

Software Engineer, Intel
Sheng is a firmware engineer working on coreboot and Slim Bootloader projects after joining Intel in 2017. Since then he has developed a keen interest in firmware development and actively involved in open source firmware initiatives . Outside of work Sheng is a tech geek and enjoys... Read More →


Wednesday May 13, 2020 1:35pm - 1:50pm
EW: Open System Firmware

1:50pm

Go Forth and Modify: Fiano
In this talk we present Fiano, Go-based tools created for manipulating UEFI images. Fiano is fast, scriptable, easy to use, and most importantly, does not require you to have UEFI source -- you can modify UEFI ROMs, remove DXEs and, if desired, replace them with a Linux kernel.
Fiano puts you in control of your systems firmware. Even if your BIOS is a blob and outside your control, these tools will help you inspect your firmware (for example, malware analysis) and improve your security.
Fiano can also improve build times, making it possible for individual DXEs to be compiled and inserted into a prebuilt image, avoiding the need to rebuild the entire firmware image.
Fiano has added preliminary support for dealing with more closed souce blobs, such is Intel FSP parsing, and limited Intel ME parsing and size reduction.
In the future, we hope to see more systems with a fully open-source firmware stack. Until such time, tools such as Fiano are necessary to give you freedom and act as a stepping stone to bring open-source into your firmware.

Speakers
RO

Ryan O'Leary

Software Engineer, Google


Wednesday May 13, 2020 1:50pm - 2:05pm
EW: Open System Firmware

2:00pm

Liquid Immersion Optimized Servers
Liquid cooling is coming to the forefront with the latest high power silicon offerings for AI and HPC workloads. Cold plate direct cooling technologies, widely used in nearly every Supercomputer offering, is straight forward, but also very complicated. Liquid Immersion techniques have potential for much simpler and higher density designs, but are held hostage to today's lower-density air-cooled systems. In this track you will learn how to tip your server design over to become optimized for life in a tank.

Speakers
avatar for Mark Shaw

Mark Shaw

GM Advanced Hardware Dev, Microsoft
Standardized, scalable hardware for the Cloud! Open CloudServers!Mark Shaw is an Architect and developer of Server Hardware for Microsoft’s Cloud+Enterprise Division. Mark and his team are working with the industry on developing the next set of server hardware.
avatar for Husam Alissa

Husam Alissa

Sr. Engineer, Microsoft
Husam is a technical lead working for Microsoft Data Center Advanced Development Team with focus on next generation technologies for hardware and infrastructure.


Wednesday May 13, 2020 2:00pm - 2:30pm
EW: Advanced Cooling Solutions

2:00pm

Designing an Efficient Programmable Switching Pipeline
The design of high-performance networks requires efficient design at all layers of the network as well as building on abstractions at each layer. This talk focuses on the device layer to show how an efficient packet processing engine can be designed and implemented without compromising on capacity or capabilities.
At the device layer, the design of high-performance switching ASICs is an optimization act on multiple dimensions. Performance, feature and scalability requirements must be balanced with latency, power and silicon area constraints. Higher level languages can be designed with a compiler driven workflow to leverage these programmable pipelines. The language for programming an ASIC based switching pipeline must focus equally on codifying packet processing flows and enabling efficient use of available hardware resources.
Broadcom has open sourced the Network Programming Language (NPL) specification in conjunction with the release of Trident 4, a 12.8Tbps fully programmable network switch ASIC. Apart from supporting basic constructs like parsers, tables and editors, NPL introduces new concepts such as “function primitives” that separate action logic from table lookups; “object strengths” that simplify resolution across various table lookups and explicit support for multiple lookup tables. More information about NPL is available at nplang.org.

Speakers
avatar for Venkat Pullela

Venkat Pullela

Distinguished Engineer, Broadcom
Venkat Pullela is a Distinguished Engineer in the Switch Architecture team at Broadcom. Prior to Broadcom, he was a Sr. Director of Engineering at Cisco leading Cat6K, Nexus 1K teams. He contributed to 8 generations of switching engines at Cisco resulting in 20+ patents and Cisco’s... Read More →


Wednesday May 13, 2020 2:00pm - 2:30pm
EW: Networking Hardware

2:05pm

Testing Firmware the DevOps Way
ITRenew is selling OCP servers under the Sesame brand, those servers come either with a UEFI BIOS or with LinuxBoot.
The LinuxBoot project is pushing the Linux kernel inside bios flash and using userland programs as bootloader.

To achieve quality on our software stack, as any project, we need to test it.
Traditional BIOS are tested by hand, this is 2020 we need to do it automatically!
We already presented the hardware setup behind the LinuxBoot CI, this talk will focus on the software.

We use u-root for our userland bootloader; this software is written in Go so we naturally choose to use Go for our testing too.
We will present how we are using and extending the Go native test framework `go test` for testing embedded systems (serial console) and improving the report format for integration to a CI.

Speakers
avatar for Julien Viard de Galbert

Julien Viard de Galbert

Senior Firmware Engineer, ITRenew France
I’ve more than 15 years of engineering experience in fields ranging from hardware design for embedded devices to embedded software and lastly system firmware (or BIOS). I’ve been focusing my career on open source since the last 8 years. I’m now working on replacing proprietary... Read More →


Wednesday May 13, 2020 2:05pm - 2:20pm
EW: Open System Firmware

2:20pm

Open Source Firmware Testing at Facebook
With LinuxBoot we became vendors of our own system firmware. In order to go to production we need a reliable quality assurance process, and firmware testing was a necessity. In this talk we are presenting ConTest (short for Continuous Testing), a modular framework aimed at automating system testing workflows, like firmware validation and provisioning. ConTest has several goals in mind: being open source and community-driven; validate as much as possible at compile time and at job submission time, to minimize unnecessary operations and run-time failures; being lightweight and infrastructure-agnostic, so it can run in Facebook’s datacenters as well as on a Raspberry Pi; being composable, thanks to an interface-and-plugins architecture; being user-oriented so that it’s not necessary to know the internals to use it effectively; and being metrics and events driven, so that users can gain valuable insights about their jobs, more than just success rate (e.g. micro-benchmarking and trend analysis).

ConTest is aimed at anyone who need to automate system-level testing. Various plugins are provided out of the box, with examples on how to use them. The users can combine them like building blocks using a simple job description format based on JSON, and test scenarios of variable complexity. When default plugins are not enough, for example in order to talk to a custom service, users can develop new plugins, and plug them just like if they were part of the core framework. Open-sourcing your own plugins is always appreciated!

Speakers
avatar for Andrea Barberio

Andrea Barberio

Production Engineer, Facebook
Andrea is a Production Engineer at Facebook, where he worked on the DNS infrastructure, cluster lifecycle automation, OS provisioning, and in the last year on open source system firmware. In his past lives, Andrea worked on the design and development of large scale network monitoring... Read More →
avatar for Marco Guerri

Marco Guerri

Production Engineer, Facebook
Marco is a Production Engineer at Facebook working on server life cycle, with a focus on platform security and open source firmware testing. Previously, Marco worked at CERN on data centre automation and capacity planning.


Wednesday May 13, 2020 2:20pm - 2:35pm
EW: Open System Firmware

2:30pm

Implementing Time Synchronization on Open Platforms
Time Synchronization capability on open networking switch is critical to next generation network infrastructure based on dis-aggregation model. From HW manufacturer’s point of view, in this workshop, we’d like to open up a bit on the time-sync design and testing, moreover, to invite community members’ (users, software developers or system integrators) involvement to tackle the last challenge together.

Speakers
MT

Milo Tseng

Director, Delta Electronics (Netherlands) B.V.
avatar for Samuel Fu

Samuel Fu

PLM, Delta Electronics


Wednesday May 13, 2020 2:30pm - 3:00pm
EW: Networking Hardware
 
Thursday, May 14
 

8:00am

LOG IN HERE for the Virtual Summit!
LOG IN HERE: https://onlinexperiences.com/scripts/Server.nxp?LASCmd=L:0&AI=1&InitialDisplay=1&ClientBrowser=0&ShowKey=86206

REGISTER HERE: https://www.opencompute.org/summit/global-summit/registration

Thursday May 14, 2020 8:00am - 4:30pm

8:59am

OCP Experience Center
OCP Accelerator Module, OAI ( Facebook, Hyve Design Solutions, Inspur, Xilinx)
OAI will demonstrate currently available OAMs as well as show functional implementations in multiple OAI systems.

OpenEDGE (Artesyn Embedded Power, ASUS, BelPower Solutions & Protection, Flex, Inventus Power, MiTAC, NOKIA)
OCP Open Edge project creates open specifications, standards and reference designs for Edge Computing and Telco use cases that require small form factor server designs that enable high performance and low latency computing. Participants of the project are across the ecosystem including Telco Service Providers, OCP solution providers, ODM/OEM manufacturers, and key technology providers. Nokia has contributed the initial Open Edge solution that creates the basis for interoperable designs.

RunBMC (Dropbox, Intel, Vertiv)
This is a demo of hardware modules following OCP RunBMC specification. The first video delivered by authors of the specification from Dropbox features the demo of HS2500 Prototype Module and HS2500 Production Module. This video also presents the standalone RunBMC Bring Up Vehicle and RunBMC-compatible Olympus chassis with Nuvoton Poleg RunBMC Module. The second video delivered by Intel presents FPGA-based RunBMC module featuring Intel® Cyclone V and Intel® MAX10. The video demonstrates the integration of this module in standard Intel 2S Skylake-X platform by using RunBMC Bring Up Vehicle card from Dropbox. In this video you can also see OpenBMC boot on Intel RunBMC module and basic management of the platform with WebUI, SOL and Remote KVM.

Open System Firmware, OSF (Facebook, Wiwynn, HP Enterprise, Intel and ITRenew).
Open system firmware is an open development project, the goal of which is to allow OCP owners to 'own their firmware' -- to move the point of control of firmware to the system owner. Starting in March, 2021, OCP badging for servers will require that systems support OSF. We are exploring multiple ways to accomplish this with components including Hostboot, UEFI, coreboot, and LinuxBoot.

OpenRMC (Inspur)
The Demo will show the establishment of OpenRMC Northbound 1.0 profile execution on top of the bundle of OpenRMC FW & HW (RMC) contributed to OCP as a complete reference solution to address the fulfilment of user's requirements/demand for a comprehensive management of an entire rack system and beyond via RESTful API through remote management portal/client console with detailed WebUI and paving the way for exploring future OpenRMC v2.0.

SEBA Solution (MITAC and EdgeCore)
SEBA solution is one of ONF project and with Edgecore and MiTAC, you can have the OCP hardware ready for SEBA.

OCP NIC 3.0 group (Amphenol, Broadcom, Dell Technologies, Facebook, Intel, Marvell Semiconductor, NVIDIA Networking - Mellanox, Quanta, TE Connectivity, UNH-IOL, Wiwynn Corporation)
The goal of this subgroup is to form a healthy OCP NIC 3.0 ecosystem through a supportive, informative and inclusive community.

96Tb DDC based Disaggregated Router ( Credo, DriveNets, Edgecore Networks, Keysight Technologies, UfiSpace )
The Open Compute Project (OCP) has selected UfiSpace, Credo and DriveNets to give the first ever live demonstration of a Distributed, Disaggregated Chassis (DDC) model as proposed to the OCP by ATT. This demonstration is the largest live demonstration of a distributed router cluster ever given in public. The technology is aligned with the vision of TIP (Telecom Infra Project) who also invited DriveNets and UfiSpace to advise on their Disaggregated Open Router Initiative and ensure its alignment with the OCP's efforts around network disaggregation.

Thursday May 14, 2020 8:59am - 4:00pm
Experience Center

8:59am

Expo Hall Open
Thursday May 14, 2020 8:59am - 4:00pm
Expo Hall

8:59am

9:00am

Storage Project Update
Jason and Dave to recap the work done in 2019, and the work ahead in 2020 and new efforts

Speakers
avatar for Jason Adrian

Jason Adrian

Storage Hardware Architect, Microsoft Azure
Jason is passionate about hyperscale and scale-out storage systems and storage architectures. His experience ranges from high-performance flash systems to archival systems, and everything in between.
DL

Dave Landsman

Dir. of Industry Standards, Western Digital


Thursday May 14, 2020 9:00am - 9:10am
EW: Storage Solutions

9:00am

Hardware Framework for Disaggregated Scale-out Firewall
Todayís service provider and large enterprise networks are expected to protect their high speed networks from security threats in order to better serve their customers. To that end, being able to incorporate security functions on the routing edge itself at high scale in a flexible and cost effective manner becomes a key criteria of success. AT&T and Palo Alto Networks are proposing a new type of OCP routing/switching white box that can also be used to host network-based firewall functionality.

Speakers
avatar for Tuan Duong

Tuan Duong

Lead Member of Technical Staff, AT&T
Tuan Duong is a Lead Member of Technical Staff at AT&T with a focus on testing and certification within the Technology Design and Architecture group for IP/MPLS Core and Edge Networks and Mobility Backhaul Transport. In his current role in AT&T labs, he has been focusing on AT&T’s... Read More →
avatar for Jim Sugg

Jim Sugg

Senior Product Manager, Palo Alto Networks
I am a senior Product Manager at Palo Alto Networks.


Thursday May 14, 2020 9:00am - 9:15am
EW: Telco & Edge

9:00am

The Journey to OCP NIC3.0 v1.0 and Beyond
OCP NIC 3.0 v1.0 is the successor to OCP Mezz 2.0. The OCP NIC 3.0 form-factor has gained strong momentum with enterprise integrators and hyper-scalers over the last two years. Major NIC and Baseboard vendors have adopted the form-factor into their product portfolios. OCP NIC 3.0 is a collaborative development effort with 14 key companies, partners and the general community.

In this workshop session, we present our specification development path from the initial drafts to the release version 1.0. We will also share the document review process, recent specification updates, key focus areas and discuss our development roadmap for beyond rev 1.0.

Future updates to the specification have much larger implications. Its critical for specification consumers be part of the review process. Input from the OCP community is key to shaping the future, therefore we look forward to your participation!

Speakers
avatar for Damien Chong

Damien Chong

Hardware Engineer, Facebook
avatar for Paul Kappler

Paul Kappler

Hardware Engineer, Intel
avatar for Thomas Ng

Thomas Ng

Hardware Engineer, Intel
Come talk to me about OCP NIC 3.0!


Thursday May 14, 2020 9:00am - 9:20am
EW: NIC 3.0 Interface/HPC

9:00am

Extending SAI beyond Switching ASIC
SAI standards have been enriched in 2019 with MACSEC, Gearbox, NAT, debuggability and more. We will talk about MACSEC and Gearbox user scenarios and SAI APIs. We will also reveal one of the latest SAI standard work in an interesting domain.

Speakers
GL

Guohan Lu

PRINCIPAL SE Manager, Microsoft
avatar for Bithika Khargharia

Bithika Khargharia

Technical Program Manager, Facebook
RH

Rita Hui

Principal Engineering Manager, Microsoft


Thursday May 14, 2020 9:00am - 9:30am
EW: Networking Software

9:00am

Sesame Fast-Start Mini-Chassis Update and Users Feedback
Assuming that many initial PoC requests cannot be processed with OpenRack because of the size and cost of the tests, last year we announced the development of a mini-rack, code-named Discovery and now available as Sesame by ITRenew Fast-Start, which can be used in any office environment. The goal is to accelerate the adoption of OCP equipment by providing a lab chassis format that can facilitate access to OCP technology. We launched the commercialization of this product in October 2019 and we would like to expose the reactions of the first buyers, as well as the new developments planned for this product.

Speakers
avatar for Jean-Jacques Chanut

Jean-Jacques Chanut

Mechanical Engineer, ITRenew


Thursday May 14, 2020 9:00am - 9:30am
EW: Rack & Power

9:00am

Entering the Era of Open Source Networking for Everyone - presented by Dell Technologies
The networking world has been quickly evolving with the disaggregation of hardware from software over the last 5 years. Emerging workloads and modern open source architectures are setting the stage for the next wave of evolution – open source networking. This session will discuss activities leading to the adoption of SONiC in Enterprise as well as how Dell EMC Networking is enabling this transformation.

Speakers
avatar for Alley Hasan

Alley Hasan

Director, Networking Business Development, Dell Technologies
A technologist by DNA, with a background in product, strategy, sales and engineering. Responsible for creating and executing on new market categories through innovation. From business case to deployment in customer environment, adept at running/incubating a startup inside the company... Read More →


Thursday May 14, 2020 9:00am - 9:30am
Executive Tracks

9:00am

Development of System Specification for UPS
Uninterruptible power supplies (UPS) are critical elements of infrastructure in the power distribution system of a data center, and as such they must remain stable as the scale of the data center grows to several MW. In the past, Facebook has experienced data center power system instability as discussed at the Open Compute Project Data Center Facility Project Call [https://www.youtube.com/watch?v=aUk_-ufhR2Y&feature=youtu.be].

The objective of this paper is to show the analytical process to establish system specifications for UPS systems to remain stable within large scale data centers, which are characterized by having a large number of regulated power supply units as rack loads. Existing UPS standards specifying performance and test requirements do not consider operation with regulated power supply units. For illustrative purposes, representative case studies are proposed and simulated in a computer program to verify the stability predictions from our analytical framework.

In the draft paper, only the bypass mode is covered, but the final paper due on Dec. 18th, 2019 will cover the other typical UPS operating modes.

Speakers
avatar for Mauricio Cespedes

Mauricio Cespedes

Power System Engineer, Facebook


Thursday May 14, 2020 9:00am - 9:45am
EW: Data Center Facility/Modular DC

9:10am

Cloud SSD - Microsoft/ Facebook Collaboration
Microsoft/ Facebook have merged their SSD requirements into a single document. This presentation will discuss the benefits of this and how this helps both system makers and SSD providers.

Speakers
avatar for Lee Prewitt

Lee Prewitt

Principal Hardware Program Manager, Microsoft
Lee has over 25 of storage industry experience having worked on many types of storage devices ranging from Magneto-Optical to CD/DVD/Blu-ray to spinning rust to Flash. Lee started at Microsoft in 2008 working in the Windows and Devices Group where he was responsible for many of... Read More →
avatar for Ross Stenfort

Ross Stenfort

Hardware System Engineer, Storage, Facebook
Ross Stenfort is a Hardware System Engineer at Facebook working on storage. He has been involved in development of SSDs, ROCs, HBAs and HDDs. He has over 40 granted patents. He has had extensive storage experience in both large and small companies including CNEX, Seagate, LSI, SandForce... Read More →


Thursday May 14, 2020 9:10am - 9:40am
EW: Storage Solutions

9:15am

Distributed Disaggregated Chassis Model Architecture
A disaggregated design architecture approach to support scalability and different size of core network

Speakers
avatar for Kei Lee

Kei Lee

Technical Director, Ufispace
Responsible for Technical Sales with merchant silicon white box hardware solution. Focus on open source architecture.


Thursday May 14, 2020 9:15am - 9:30am
EW: Telco & Edge

9:20am

OCP NIC 3.0 Compliance Program Update
To eliminate headaches caused by unknown or underdeveloped devices and ensure system interoperability in large deployments the OCP NIC 3.0 WG is finalizing a compliance/logo program to provide users and integrators with the confidence that products implementing the OCP NIC 3.0 form factor will meet the specification requirements. Compliant devices will be published on an integrators list and receive a logo to denote demonstrated compliance and interoperability; UNH IOL will lead the certification efforts. Supported by system integrators and hyperscale datacenter users, wide adoption of the program will reduce issues found in the field, and promote the benefits that OCP NIC 3.0 technology has to offer.

Speakers
MK

Michael Klempa

Technical Manager, Ethernet and Storage, UNH-IOL
avatar for David Woolf

David Woolf

Senior Engineer Datacenter Technologies, UNH-IOL
David Woolf currently leads Data Center initiatives at the University of New Hampshire InterOperability Laboratory (UNH-IOL).


Thursday May 14, 2020 9:20am - 9:35am
EW: NIC 3.0 Interface/HPC

9:30am

NOS for High Density 10G PON OLT
With the recent anouncement made by Broadcom regarding the new high-density PON OLT MAC, with samples already available for "qualified customers", it is likely we are going to see possibly during 2020 OCP Summit the first products powered by those devices. Diferent from the first generation (2016), when the future of this technology was not so clear and the designs were not competitive, we are going to see real commercial solutions taking over the market.
On the software side, there is still a gap. ONF has been promoting the disaggregation of the software side through the VOLTHA Project. However, while VOLTHA does not reach the necessary maturity and completeness, all the OLT design contributions will remain useless until a solid NOS alternative becomes available in the market, open source or commercial.
We are going to show how we are combining open source and commercial solutions under a modular architecture that can speed up the adoption of OCP OLT designs and its deployment in scale.

Speakers
avatar for Ricardo Pianta

Ricardo Pianta

Engineering Manager, Datacom - Teracom Telematica
Besides his main responsability leading the R&D team at Datacom, Ricardo enjoys his business development role and represents the company in a number of foruns like ONF, OCP and TIP. He became a huge enthusiat of disaggregation initiatives and a very active member of the community... Read More →


Thursday May 14, 2020 9:30am - 9:45am
EW: Telco & Edge

9:30am

SONiC 2020 Update and Beyond
SONiC has been evolving significantly since 2019 summit. It has been extended to much richer scenarios outside Cloud and Data Centers, e.g. Edge, Security, Enterprise, Management. We will look into two areas in this talk and demo how SONiC provides a scalable, high performance solution for the new scenarios.

Speakers
avatar for Xin Liu

Xin Liu

Principal Product Manager, Microsoft Azure, Microsoft
LY

Lihua Yuan

PARTNER SOFTWARE ENG MGR, Microsoft
Lihua has been with Microsoft for 10 years, working on building hyper scale network since 2008. He got his Ph.D in Computer Science from UC Davis. He has around 11 publications, and a rich experience in speaking on all sorts of conferences, from Sigcomm, to INFOCOM, OCP, ONS etc... Read More →


Thursday May 14, 2020 9:30am - 10:00am
EW: Networking Software

9:30am

Battery Backup Solutions - Options; Advancements; and Safety
As the expansion of cloud and edge computing continues at a rapid pace, many questions arise around system architecture and the optimal user experience. With high performance as a baseline expectation, system security and data integrity are also “must haves.” A proper battery backup solution ensures system reliability, data integrity and, in certain instances, extended run-time during short term power interruptions. A design with enough run time for a graceful shutdown helps ensure security and easy launching once the regular power is restored. The system architect must also choose from a variety of options when considering battery backup. Will power come from a full system UPS or be integrated in each server or data storage blade? Or, will there be a power shelf tied in with the PSU? Will the battery power be regulated or not? There is also an industry wide question about the safety of Li-ion in datacom and telecom systems, as well as new agency certifications and fire codes now part of the equation. How do we address these new and valid concerns? This session will help articulate the traditional system level options– as well as provide an introduction to some of the critical design parameters for the battery system itself. Topics will include: 1. Thermal Monitoring and Management 2. Cell Selection a. Chemistry options b. New cell sizes 3. Safety Considerations and Techniques a. Voltage levels b. Thermal propagation within pack 4. New Agency Certifications / Requirements

Speakers
avatar for Howie Granat

Howie Granat

Director - Business Development, Commercial, Industrial Consumer, Inventus Power
As BD Director, Howie is responsible for the strategy and business development activities within the critical power industry segment and is the program lead for all OCP related projects. He has a thirty-year history in the electronics industry with a distinct focus on datacom and... Read More →
avatar for Ilyas Ayub

Ilyas Ayub

VP - Engineering, Inventus Power
As the VP of Technical Center Americas, Ilyas oversees Product Development, Application and Project Engineering teams who are responsible for the design of highly complex batteries, chargers and power supplies. He is a skilled engineer with over 15 years of electronics experience... Read More →


Thursday May 14, 2020 9:30am - 10:00am
EW: Rack & Power

9:30am

Supermicro: Delivering Open yet Secure and Cutting Edge Compute, Storage and Networking Solutions
In this session, the presenter will lead the audience through the maze of demanding requirements placed by the applications on today’s infrastructure and how to design & architect to the security, integration and open requirements of a flexible and scalable solution for tomorrow .

Speakers
avatar for Sim Upadhyayula

Sim Upadhyayula

Senior Director Solutions Enablement & FAE, Supermicro


Thursday May 14, 2020 9:30am - 10:00am
Executive Tracks

9:35am

TX / RX Conformance Testing with NIC 3.0
Since 1988, the UNH InterOperability Laboratory has enabled multi-vendor interoperability, conformance testing and developed custom testing solutions while preparing students for careers within the computer and engineering industries. Over the years we have grown steadily into one of the premier independent labs for new technologies, allowing us to become a thought leader within the data and networking world. We have recently been working hard to bring full testing capabilities for the OCP NIC 3.0 to our lab in preparation for the OCP NIC 3.0 certification program coming out early next year. This technology and certification program will be essential to improving hyperscale data centers and HPC deployment in the near future. NIC 3.0’s capabilities and footprint will help to streamline large scale operations while allowing them to stay cool, making them a key part of future deployments. An essential part of the compliance program is ensuring PCIe 3.0 and 4.0 TX / RX conformance for the NICs. We have created multiple testing suites around the new NIC 3.0 hardware needed to complete this testing in an easy and efficient manner. This includes a detailed look into the differences in S-Parameters between a traditional PCIe CBB and CLB and our NIC 3.0 CBB and CLB. We will also go into how this testing capability can be integrated into vendor’s design process to help eliminate PCIe TX / RX communication errors earlier in development.

Speakers
AG

Alex Girard

PCIe Test Engineer, UNH-IOL
MB

Michael Baumer

PCIe Technician, UNH-IOL


Thursday May 14, 2020 9:35am - 9:45am
EW: NIC 3.0 Interface/HPC

9:40am

Enabling Full Stack QoS with IO Priority
The HDD IO priority has been explored on Facebook storage server Bryce Canyon. The feature can help enable QoS at IO level and provide a methodology to better manage latency for our workload. However, the full stack implementation requires standardization efforts of creating innovative communication channel between the host and device on defining target metrics and policies. The communication syntax needs to be consistent across vendors, capacity and SAS SATA protocols. In this talk, we would like to present potential latency management benefits based on the preliminary experiments that has been done on our Bryce Canyon platform. We will also discuss about some certain challenges we are facing such as number of priority levels, SAS & SATA difference and tail latency management etc. Those challenges cannot be easily resolved without the collaboration across the industry.

Speakers
avatar for Yong Jiang

Yong Jiang

Storage Engineer, Facebook
AS

Abhinav Sharma

Software Engineer, Facebook


Thursday May 14, 2020 9:40am - 10:10am
EW: Storage Solutions

9:45am

Open Rack Deployments in Telco Data Centers
This presentation will share the details for Nokia Open Rack modular configurations and tool used to configure racks. Nokia Build to Order model (BtO) allows user to configure server and storage configurations from validated components that are available in Nokia configurator. Server and storage configurations can be used with available rack layouts that fits best for users requirements. Tool will generate detailed asset file about completed configuration that integration factory will use to build the rack for delivery. Servers, storage devices and cablings are all factory integrated and tested after which system is packed and delivered to customer site with turn key mode. Customer need to power up the rack and use Nokia data center manager (NADCM) to upload device MAC addresses and IP plan from asset file and start automated deployment for the Hardware configuration.

Speakers
avatar for Esa Hietto

Esa Hietto

Product Architect, Nokia


Thursday May 14, 2020 9:45am - 9:55am
EW: Telco & Edge

9:45am

OCP NIC 3.0 Manageability Features Overview
OCP NIC 3.0 specification has enabled a wide spectrum of next generation NIC board designs. One of the key aspects of OCP NIC 3.0 specification is standards-based manageability. In this talk, we will provide an overview of OCP NIC 3.0 manageability features that are based on a number of DMTF standards including NC-SI, MCTP, and PLDM. We will cover areas including sideband management interfaces, MAC address provisioning, temperature reporting, power consumption reporting, NC-SI package addressing/hardware arbitration, firmware inventory, firmware update, and FRU content.

Speakers
avatar for Hemal Shah

Hemal Shah

Distinguished Engineer, Broadcom
Hemal Shah is a Distinguished Engineer and Systems/Software/Standards architect in the Compute and Connectivity (CCX) division at Broadcom Inc. He leads a team of architects. In his role, he is responsible for the product architecture definition of Ethernet NIC products, NIC integration... Read More →
avatar for Patrick Caporale

Patrick Caporale

Distinguished Engineer, Lenovo


Thursday May 14, 2020 9:45am - 10:00am
EW: NIC 3.0 Interface/HPC

9:45am

Data Center Electrical Specification for System Stability
This paper presents a set of specifications and design guidelines to ensure data center electrical system stability and to reduce voltage unbalance. Facebook experienced electrical stability and resonance at multiple data centers and identified that the impedance interaction between the building electrical system and the power supplies are the root cause for the instability. Higher building impedance drives the power supplies to instability; therefore, it is critical to keep the impedance the electrical system under a certain limit. This paper focuses on defining inductance limits for the passive electrical components including switchboards, reactors, transformers and busways. The goal is for equipment vendors to comply with the limits to make sure system stability and to avoid high system voltage unbalance by design. Impedance-based power stability analysis and generalized Nyquist stability criterion are used to define the limits given on the specification. Field impedance measurements are presented to demonstrate the stability limit on selected data centers.

Speakers
avatar for Melaku M. Mihret

Melaku M. Mihret

Electrical Systems Engineer, Facebook
I am Melaku Mihret, Electrical System Engineer at Facebook.


Thursday May 14, 2020 9:45am - 10:30am
EW: Data Center Facility/Modular DC

9:55am

OpenEDGE Base Specification & Sled Concepts
EDGE requirements are driving a need for more flexible and versatile solutions. In order to address these demanding requirements in a quick and efficient manner, new and innovative OpenEdge sled concepts must be realized. We will demonstrate through this session different sled concepts provisioning for compute, storage, switching, and I/O to meet the challenging needs of telco, IOT and other market segments.

Speakers
avatar for Dirk Blevins

Dirk Blevins

Telco Solution Architect, Intel
My name is Dirk Blevins. I am a Platform Architect at Intel and have been with the company for 18 years. My previous experience was in building printers for Lexmark Corporation and Military and Defense with Harris Corporation. I have done every job function imaginable in engineering... Read More →


Thursday May 14, 2020 9:55am - 10:15am
EW: Telco & Edge

10:00am

PLDM Modeling of OCP 3.0 NIC
PLDM for Monitoring and control (DSP0248) was defined by DMTF as the management protocol for in-box devices. OCP 3.0 allows for multiple configurations to be used. A PLDM model is presented, allowing managing the different card types, using a consistent modeling scheme based on DMTF DSP2054. This talk covers the different elements in the PLDM schema, their enumeration method and how to use the same schema to address all the different card types defined in OCP 3.0 spec

Speakers
YI

Yuval Itkin

Distinguished architect, Mellanox
Servers management and OCP server hardware.
BW

Ben Wei

System Software Engineer, Facebook


Thursday May 14, 2020 10:00am - 10:15am
EW: NIC 3.0 Interface/HPC

10:00am

SONiC: Multi-Tenant Data Centers with EVPN VXLAN
SONiC is rapidly emerging as the next generation Network OS of choice for data center deployments. This community driven, opensource Network OS has been deployed in various webscale datacenters. This session will talk about how SONiC is now expanding beyond the webscale data centers to multi-tenant agile data centers for Service Providers and Enterprises with support for virtual networking and programmatic automation and monitoring. This talk will describe a new SONiC feature initiative - EVPN over VXLAN. This technology allows easy configuration and management of private L2 and L3 Virtual Network Instances on a shared IP underlay fabric, opening up a variety of network virtualization use cases. The talk will cover the target use cases, the feature functionality, and the SONiC design.

Speakers
avatar for Adam Yeung

Adam Yeung

Software Engineering Manager, Broadcom
Adam Yeung is a network software veteran, inventor of 9 US Patents. At Broadcom, Adam is leading the network OS data plane design and development for EVPN VxLAN, Multi-chassis Dual Homing, L2, ACL services, and IP Multicast.
avatar for Kamini Santhanagopalan

Kamini Santhanagopalan

Product Line Manager, Broadcom
Accomplished and Passionate Product Line Manager with 11+ years of progressive product management, technical marketing and solutions engineer experience spanning Enterprise Data Center and Campus Switching technologies.


Thursday May 14, 2020 10:00am - 10:30am
EW: Networking Software

10:00am

Open Rack Frame
The Open Rack V3 frame is currently in development. This will be an update on the progress the team is making on the Open Rack V3 Frame design.

Speakers
avatar for Steve Moore

Steve Moore

Project Engineer, Rittal
Contact Information: Steven Moore Project Engineer Phone: +44 01752 207 654 Email: smoore@rittal-csm.co.ukI have worked for Rittal over the past year and a quarter, starting out as a Design Engineer and transitioning to a Project Engineer. I am now the Project Engineer of one of Rittal’s... Read More →
avatar for Glenn P. Charest

Glenn P. Charest

Technical Leader, Facebook


Thursday May 14, 2020 10:00am - 10:30am
EW: Rack & Power

10:00am

New Era of Software defined Networking and Storage Infrastructure - presented by Xilinx
Software Defined Networking and computational storage is transforming Data Center infrastructure in several profound ways.  As compute migrates out to the NIC and where the data lives in storage, a whole new set of domain specific architectures are being created and deployed.   In this session we will present Xilinx’s vision regarding this important new trend and explain how FPGA technology will play an important and critical role.

Speakers
avatar for Jamon Bowen

Jamon Bowen

Product Marketing Director, Xilinx
Jamon was an early pioneer in the enterprise flash storage market, setting the requirements and bringing the first enterprise flash array to market in 2007 while at Texas Memory Systems (now IBM).  He has held board positions at SNIA for the solid-state storage initiative and is... Read More →


Thursday May 14, 2020 10:00am - 10:30am
Executive Tracks

10:10am

HDD Queuing for Cloud Configurations
When hard disk drives are use in cloud environments, the effects of each drive's queuing methods cease to be limited to the interactions between that one drive and the host. Instead, individual drives become participants in a larger management algorithm for data accesses. 

Speakers
avatar for Ralph Weber

Ralph Weber

Technologist, Western Digital
Ralph Weber edited the SCSI Primary Commands standard from is inception in 1994 until 2016, coined the acronym CAP (Commands, Architecture, and Protocols) as the moniker for the most active SCSI Working Group, chaired T10 Plenaries for two years ending in 2016, and has been honored... Read More →
avatar for Bill Boyle

Bill Boyle

Fellow, Western Digital
Bill Boyle has been in the hard disk drive industry for 30 years (25 at Western Digital). He has a breadth of experience in systems architecture – HDD controller firmware, servo firmware, systems storage interfaces – and has accumulated a considerable aptitude for designing widely... Read More →


Thursday May 14, 2020 10:10am - 10:40am
EW: Storage Solutions

10:15am

OCP 3.0 NIC Monitoring and Management in OpenBMC
In this talk we will go over OpenBMC NIC management features on OCP Yosemite V3 platform. In this platform there's a single multi-host NIC connecting to a single BMC and 4 servers. NIC management for OpenBMC is done through a user space monitoring daemon and a command line tool to communicates with NIC over NC-SI. One of its main responsibility is to monitor NIC link status and performs automated recovery and remediation as needed. Additionally, it monitors NIC events through subscribing to various NC-SI AENs (standard and OEM) and triggers external logging and remediation as necessary. I will also give a an overview of command line tool in OpenBMC to aid NIC debugging.

Speakers
BW

Ben Wei

System Software Engineer, Facebook


Thursday May 14, 2020 10:15am - 10:25am
EW: NIC 3.0 Interface/HPC

10:15am

Guarding Assets in Edge Platforms
A discussion of various aspects and unique challenges of Security at the Edge.

Speakers
DA

Dr. Ahmad Atamli

Chief Security Architect, Nvidia


Thursday May 14, 2020 10:15am - 10:35am
EW: Telco & Edge

10:25am

Design Guidelines for OCP NIC 3.0 Secure Firmware
Device security is becoming increasingly important in today’s data centers. Secure firmware is a recommended feature in OCP NIC 3.0 specification. In this talk, we will discuss design and implementation guidelines for secure firmware. We will cover areas including signed firmware, secure boot, secure firmware update, root-of-trust, device attestation, and firmware measurements.

Speakers
avatar for Hemal Shah

Hemal Shah

Distinguished Engineer, Broadcom
Hemal Shah is a Distinguished Engineer and Systems/Software/Standards architect in the Compute and Connectivity (CCX) division at Broadcom Inc. He leads a team of architects. In his role, he is responsible for the product architecture definition of Ethernet NIC products, NIC integration... Read More →
avatar for Eric Spada

Eric Spada

Distinguished Engineer, Broadcom


Thursday May 14, 2020 10:25am - 10:40am
EW: NIC 3.0 Interface/HPC

10:30am

Colo Facility Guidelines and OCP Ready Assessment Revision
The OCP Foundation developed guidelines and the requirements and process for a colocation data center facility provider to become a listed Colo Solution Provider on the OCP Marketplace. The facility provider must perform a self-assessment of their facility using the site assessment scorecard, which is presented to and reviewed by the Data Center Facility Project Community and the OCP Incubation Committee. The facility provider must also meet all requirements set forth by the OCP Foundation.

It has been 12 months since the first version of the 'Colo Facility Guidelines for the Deployment of Open Compute Racks' was published and the OCP Ready TM site assessment requirements were finalised. This presentation will provide an update on the work to revise the Colo facility guidelines document and OCP Ready site assessment so that it is in alignment with the latest OCP technology designs i.e. Open Rack V3.0 and the requirements for Advanced Cooling Solutions etc.

Speakers
avatar for Mark Dansie

Mark Dansie

Data Center Transformation Specialist, OCP
Mark Dansie has been involved in the computing and IT infrastructure services industry for over 36 years and is the OCP Ready TM Lead. His career experience in consultancy, project and programme management, spans all aspects of building, campus and data center infrastructure, delivering... Read More →


Thursday May 14, 2020 10:30am - 11:00am
EW: Data Center Facility/Modular DC

10:30am

Open Routing in the Data Center
As more people come online and consume richer content, the complexity of the networks underlying the flow of information also grows. Unlike network hardware which evolved multiple times in a decade, routing protocols haven’t got much traction. For non-standard complex network, the need for routing solutions arises. New advances in routing often require carefully extending an existing protocol or designing new overlay extensions.
In this talk Saif presents Open Routing Platform, Open/R. A routing platform - designed, developed and open-sourced by Facebook. It started with a need for routing on Terragraph, wireless mesh networks, then has evolved and adapted into Facebook’s Data-Center and Wide Area Networks. Talk emphasizes the use-cases and share experience of routing evolution from BGP to Open/R routing in Facebook Data Center network. We also intend to share technical details on architecture of protocol, design principles that went into it, and critical lessons learned from operating Open/R in our production networks.

Speakers
avatar for Saif Hasan

Saif Hasan

Software Engineer, Facebook
Saif Hasan is software engineer at Facebook. Primarily works on building software systems to control the networks. Currently working on routing protocols, previously worked on network visualization and Egress SDN controller.


Thursday May 14, 2020 10:30am - 11:00am
EW: Networking Software

10:30am

Overview of Open Rack V3 Power System
The presentation will go over the progress being made by several partner companies on development of Open Rack V3 Power System.

Speakers
avatar for Shaun Harris

Shaun Harris

Director Power Systems and Compliance, Microsoft
avatar for Tiffany Jin

Tiffany Jin

Mechanical Engineer, Facebook
avatar for Hamid Keyhani

Hamid Keyhani

Technical Lead, Facebook
BS

Banha Sok

Microsoft


Thursday May 14, 2020 10:30am - 11:00am
EW: Rack & Power

10:30am

Accelerated Computing Deployment Made Easy over Open Networking Solutions - presented by NVIDIA Networking – Mellanox
As Moore’s Law has slowed, traditional CPU and networking technologies are no longer sufficient to support the emerging workloads like machine learning, AI security or video over IP. The presentation will go over various use cases and demonstrate how easy it is to design and deploy today, Accelerated Computing systems over open networking technologies, which boost application performance in both, bare-metal and virtualized infrastructure.

Speakers
avatar for Motti Beck

Motti Beck

Sr. Director, Enterprise Market Development, NVIDIA Networking – Mellanox
Motti Beck is Senior Director Enterprise Market Development at NVIDIA Networking – Mellanox. Before joining Mellanox Motti was a founder of BindKey Technologies an Electronic Design Automation startup that provided deep submicron semiconductors verification solutions and was acquired... Read More →


Thursday May 14, 2020 10:30am - 11:00am
Executive Tracks

10:35am

OCP Open Edge Switch Contribution
In this presentation we will describe the requirements for a NEBS compliant Open Edge switch and provide details about the design logic applied for the Open Edge switch implementation. Contributed design specification complies with OCP tenets (efficiency, scalability, openness, impact) and OCP open edge specifications. In addition to hardware, this presentation will also give an introduction to the switch functionality implementation that uses SONiC network operating system as basis.

Speakers
avatar for Mika Hatanpaa

Mika Hatanpaa

Head of R&D, Nokia
Mika is an experienced and passionate Head Of R&D with strong knowledge and experience in Telecommunications, Data Center Infrastructure, Cloud Infrastructure, Cloud Computing, Edge Cloud and Radio Access Networks, including 5G. Mika has proven track record in leading world class... Read More →


Thursday May 14, 2020 10:35am - 10:55am
EW: Telco & Edge

10:40am

Multi-Host Design Considerations for OCP NIC Implementations
OCP NIC 3.0 specification enables an implementation of a NIC that is shared by multiple host complexes. A multi-host NIC implementation needs to take into account design considerations including PCIe bi-furcation, host isolation, QoS, buffering, power management, resets, and firmware update. In this talk, we will discuss these design considerations for multi-host NIC implementations.

Speakers
avatar for Hemal Shah

Hemal Shah

Distinguished Engineer, Broadcom
Hemal Shah is a Distinguished Engineer and Systems/Software/Standards architect in the Compute and Connectivity (CCX) division at Broadcom Inc. He leads a team of architects. In his role, he is responsible for the product architecture definition of Ethernet NIC products, NIC integration... Read More →
avatar for Damien Chong

Damien Chong

Hardware Engineer, Facebook


Thursday May 14, 2020 10:40am - 10:55am
EW: NIC 3.0 Interface/HPC

10:40am

NVMe HDD's - The Future Starts Now
The SAS and SATA interfaces have dominated the storage market for nearly two decades, but is this the right interface for the future? For SSDs, the market is already shifting to NVMe for its higher bandwidth, lower latency, and overall light-weight interface. For the HDD market, SAS and SATA are plenty fast for many more years. Why would we consider NVMe? Join us as we explore some of the un-tapped opportunities with NVMe HDDs.

Speakers
avatar for Jason Adrian

Jason Adrian

Storage Hardware Architect, Microsoft Azure
Jason is passionate about hyperscale and scale-out storage systems and storage architectures. His experience ranges from high-performance flash systems to archival systems, and everything in between.


Thursday May 14, 2020 10:40am - 11:10am
EW: Storage Solutions

10:55am

Next Generation of Multi-Host NICs
Multi-Host NIC allows significantly increased compute density and infrastructure efficiency. But it also poses challenges for deployment at scale. These challenges involve various aspects of the design and operation like network congestions, flexible QoS, host isolation, power management, PCIe Bi-furcation, resets and software updates.
In this talk we would discuss some of the challenges experienced during deployment and how these shaped next generation design of Multi-Host NICs.

Speakers
avatar for Elad Wind

Elad Wind

Sr. Director, Solution Engineer, Mellanox
avatar for Rakesh Hemnani

Rakesh Hemnani

System Engineer, Facebook


Thursday May 14, 2020 10:55am - 11:10am
EW: NIC 3.0 Interface/HPC

11:00am

What Defines a Modular Data Center? (Open Discussion)
The tenents of the OCP (Efficiency, Scale, Openness, & Impact) are in great alignment with the base principals of a modular approach to data center construction. However, what defines a modular data center (MDC) is still up to debate. In this session, the community will be invited to share with their experience of MDCs, their expectations of MDCs, and the problems they believe a modular data center could solve.
Participants will share and help shape the direction of the MDC subgroup's direction for future contributions based on real life needs and possibilities. Our focus will be to maintain the efforts of making the data center more efficient, flexible, and scalable. We will work together to break down proprietary IT infrastructure to achieve greater choice, customization, and cost savings.

Speakers
avatar for Rob Coyle

Rob Coyle

Director of Sales and Marketing, PCX Corporation
As the son, grandson, and brother of carpenters, my education in construction began early. This background, combined with my passion for innovative technology solutions, led me to modular construction. The construction industry has been historically slow-to-change, yet the problems... Read More →


Thursday May 14, 2020 11:00am - 11:30am
EW: Data Center Facility/Modular DC

11:00am

SONiC Troubleshooting Experiences at LinkedIn
Based on LinkedIn deployment experiences, will present several real troubleshooting cases and how we root cause the issues and what we learnt from those cases.

Speakers
ZX

Zhenggen Xu

SW Enginneer, Linkedin


Thursday May 14, 2020 11:00am - 11:30am
EW: Networking Software

11:00am

Deep Dive on Open Rack V3 Power Shelves
2RU Shelf (Mechanics, Busbar Connection) and the Current Share between the units - Ralf Pieper, Director of Engineering, Delta Energy Systems

1OU Shelf Architecture - Harry Soin, Senior Director Technical Marketing, Artesyn Embedded Technologies
Overview (Top Level Specs)
Mechanical Design
Thermal Management
Scalability and Control Options

Rectifier and SMC Design - Lam Vu, Lite-On Technologies

Speakers
avatar for Harry Soin

Harry Soin

Senior Director Technical Marketing, Artesyn Embedded Power
Harry Soin is the Senior Director of Technical Marketing at Artesyn Embedded Power and is responsible for the Hyperscale segment. Harry has been working in this capacity since Jan 2016 in San Jose,CA. Prior to this Harry managed the China Design Engineering facility for Artesyn... Read More →
LV

Lam Vu

Director of Engineering, LITE-ON
avatar for Ralf Pieper

Ralf Pieper

Engineering Director, Delta Electronics
Ralf Pieper is the R&D Director for Server, Networking and Telecom applications and he works for Delta Energy Systems (Germany) since 1998. He studied electrical engineering at the University of Paderborn, Germany and after he graduated his Master degree, he continued to work for... Read More →


Thursday May 14, 2020 11:00am - 11:30am
EW: Rack & Power

11:00am

Panel Discussion: Disaggregate the Data Center & Conquer Complexity - Presented by Hyve
There is increasing variety in hyperscale workloads, and in the types of computation and acceleration engines to address them. These engines need to operate on large scales of data and do so quickly, leading to unprecedented engineering and planning challenges in hyperscale datacenters. One efficient way to address this complexity is to disaggregate the various subsystems in the server and distribute them in the data center. The concept is known but has been difficult to execute in the past. But we may finally be ready to take this on. Our panel has extensive experience in architecting and deploying hyperscale infrastructure and are subject matter experts in servers, storage and networking, so we will cover the key issues in disaggregating data centers and how to address them.

Moderators
avatar for Jay Shenoy

Jay Shenoy

VP of Technology, Hyve Solutions
I drive the road map and product planning for Hyve Solutions. Talk to me about any of your digital infrastructure needs!

Speakers
avatar for Anjaneya (Reddy) Chagam

Anjaneya (Reddy) Chagam

Senior Principal Engineer, Intel
Anjaneya “Reddy” Chagam is a Senior Principal Engineer and Chief Software-Defined Storage Architect in Intel’s Data Center Group. He is responsible for developing software-defined storage strategy, architecture, and technology initiatives. He is a member of the Technical Steering... Read More →
avatar for Manoj Wadekar

Manoj Wadekar

Storage Hardware Architect, Facebook
Current Focus: Solve ambiguous problems for hyper-scale infrastructure. Find the problems, drive innovative solutions.Background:28+ years in Server, Storage, and Networking industry.Specialties: Transforming hyper-scale infrastructure to drive performance and efficiency (proven for... Read More →
avatar for Parantap Lahiri

Parantap Lahiri

VP NW and DC Engineering, eBay
Parantap currently leads network, datacenter and server hardware engineering at eBay as a vice president. In past, Parantap led network engineering and architecture at Microsoft for several years before his years at Juniper Networks where he led solution engineering for Contrail SDN... Read More →
avatar for Amir Michael

Amir Michael

Chief Technology Evangelist, Lightbit
avatar for Jai Menon

Jai Menon

Chief Scientist, Fungible


Thursday May 14, 2020 11:00am - 12:00pm
Executive Tracks

11:10am

Managing a Socket Direct OCP NIC 3.0
OCP NIC 3.0 introduces standard support for multi socket servers. When a multi-socket server with multiple PCIe connections operates a NIC it requires some special management features. This talk discusses the implications of operation in Socket-Direct mode on Inventory management from the BMC as well as the IT-related pre-boot environment configuration considerations.

Speakers
YI

Yuval Itkin

Distinguished architect, Mellanox
Servers management and OCP server hardware.
PS

Patrick Schoeller

Master Technologist, HPE
I have actively conttributed to our industry for 35 years, from simple database management to embedded management of Industry Standard (computer) Servers. I am active in the DMTF standards body and a contributer to the OCP 3.0 NIC specification. I lead small development teams around... Read More →


Thursday May 14, 2020 11:10am - 11:25am
EW: NIC 3.0 Interface/HPC

11:10am

Ethernet Attached NVMe-oF Devices and Platforms
As the NVMe-oF ecosystem continues to mature, storage systems now have design choices for the type of external and internal fabrics to use, as well as the various attach points. Several Ethernet-based protocols (RoCE, iWARP, TCP) have emerged as design choices for storage fabrics and interfaces. There are unique properties and trade-offs associated with those choices, as well as different implementation and acceleration paths. An industry discussion has ensued on how close to the storage the Ethernet I/F should be carried: data center, rack, or even device. This presentation will provide a broad view of Ethernet-attached NVMe-oF disaggregated storage systems and what it would take for successful wide deployment of those systems.

Speakers
avatar for Ihab Hamadi

Ihab Hamadi

Fellow, Datacenter Systems, Western Digital
Ihab Hamadi is Sr. Director of Engineering at Western Digital, where he focuses on the innovation and growth of the company's platforms business, delivering data and application-centric solutions for the enterprise and cloud.  He has a solid 21-year track record of building leading-edge... Read More →


Thursday May 14, 2020 11:10am - 11:40am
EW: Storage Solutions

11:15am

Building the Open Source Edge with the Arm Ecosystem
Today's networks are evolving towards a future where a trillion connected devices will be generating data that needs to be managed. The rise of Edge and 5G will require solutions that can provide not only latency and bandwidth but also power efficiency and flexibility. The Arm ecosystem is uniquely situated to provide these solutions.

The talk will discuss the work Arm and its partners are doing with the open source community, looking at both open source software and hardware initiatives. We will discuss Arm's leadership in the Akraino Edge Stack project to provide a flexible open-source framework for Edge as well as blueprints for specific use cases. We will present the Software Enabled Broadband Access (SEBA) for wireline connectivity, and the Radio Edge Cloud (REC) integrating the RAN Intelligence Controller (RIC) under the Telco Appliance Blueprint Family using OCP openEDGE as the hardware.

Arm ecosystem partner Ampere computing will describe their solutions for SEBA and REC based on their half-width OCP openEDGE compute sled based on the Ampere eMAG 32 core processor. Ampere has contributed this openEDGE platform to the OCP community. The compute sled is available today in the OCP marketplace through Wiwynn, Mitec & other partners.

Speakers
avatar for Tina Tsou

Tina Tsou

Enterprise Architect, Arm
Tina Tsou is an innovator and a visionary with far-reaching accomplishments within the technical engineering realm. As Arm’s Enterprise Architect, Tina serves in the highly visible Technical Lead role for the Enterprise Open Source Enablement team, where she analyzes, designs, and... Read More →
avatar for Amar Dhamdhere

Amar Dhamdhere

Sr. Director, Product Management, Ampere Computing
Amar is a Sr. Director of Product Management at Ampere Computing responsible for delivering compute products for the Cloud and Edge market. Let us discuss how Ampere solutions fit in your Telco Edge and Cloud applications.


Thursday May 14, 2020 11:15am - 11:30am
EW: Telco & Edge

11:25am

PCIe Technology: Connecting OCP Architectures of the Future
PCI Express® (PCIe®) architecture has enabled high performance in various systems and platforms for almost thirty years, and PCI-SIG® plans to continue this forward trajectory. While all PCIe specifications are backwards compatible, the evolution of PCI-SIG’s I/O technology is paving the way for future systems to make their mark on the high-performance computing (HPC) industry. PCIe specifications consistently deliver enhanced performance, unprecedented speeds and low latency. The PCIe 4.0 and PCIe 5.0 specifications—reaching 16 GT/s and 32GT/s transfer rates respectively—laid the groundwork for the PCIe 6.0 specification. PCIe 6.0 technology will continue the tradition of doubling previous data rates at 64 GT/s and rising to meet the needs of the most demanding workloads. The current PCI-SIG specification roadmap from the PCIe 4.0 specification to the PCIe 6.0 specification will continue to enable OCP architectures and platform designs for years to come.

In this session, attendees will:
• Examine case studies where PCIe specifications are being deployed in OCP architectures.
• Learn about the development status and exciting feature in the PCIe 6.0 specification, targeted for completion in 2021.
• Receive an update on official PCIe 4.0 specification compliance testing and current industry adoption.
• Gain insight into PCIe 5.0 specification traction and a status update on the development of the compliance tests.

Speakers
avatar for Al Yanes

Al Yanes

PCI-SIG Chairman and President, PCI-SIG
Al Yanes has served as president of the PCI-SIG since 2003 and chairman since 2006 and is a Distinguished Engineer for IBM in the Systems & Technology Division. He has 26 years of experience working with ASIC design in the I/O industry. Yanes holds 25 patents for PCI™ and other... Read More →


Thursday May 14, 2020 11:25am - 11:40am
EW: NIC 3.0 Interface/HPC

11:30am

Power for openEdge Applications
Edge Computing is a critical technology enabling applications requiring low latency compute and storage such as IoT and Autonomous Vehicles.
The Open Edge project within OCP provides specifications for critical interfaces, system performance and environmental conditions.
System power is provided by two power supplies that can accept either traditional Telco -48Vdc, or wide-range AC input.
Open Edge equipment must have seamless performance regardless of the input voltage.
We will highlight the key features that enable this, and the benefits to customers such as service providers.

Speakers
avatar for Jim Nelson

Jim Nelson

Dir. Technical Marketing, Flex
Jim is part of the FlexPower team within the $26B Flex corporation. FlexPower specializes in providing Sketch-to-Scale TM solutions to Communications, Data Center and Enterprise customers. FlexPower is the 7th largest merchant power supplier. Specialties included rack-mounted equipment... Read More →


Thursday May 14, 2020 11:30am - 11:45am
EW: Telco & Edge

11:30am

Open Rack V3 Questions
Speakers
avatar for Caleb Lusk

Caleb Lusk

Engineering Systems Consultant, Rittal North America
avatar for Steve Mills

Steve Mills

Technical Lead, Facebook
OCP Incubation Committee Member
avatar for Hamid Keyhani

Hamid Keyhani

Technical Lead, Facebook


Thursday May 14, 2020 11:30am - 12:00pm
EW: Rack & Power

11:30am

90KW MDC Design Package Submission
This presentation will be to share the presentation about the design files now made open source.

PCX has developed a 90KW MDC to meet the OCP 90KW specification. These design files have been accepted by the IC. [*IC approval pending] In this session the participants will learn the driving factors for this design and problems it solves. Additionally, the participants will be given the opportunity to provide feedback that will drive future design versions made available to the community.

Speakers
avatar for Rob Coyle

Rob Coyle

Director of Sales and Marketing, PCX Corporation
As the son, grandson, and brother of carpenters, my education in construction began early. This background, combined with my passion for innovative technology solutions, led me to modular construction. The construction industry has been historically slow-to-change, yet the problems... Read More →


Thursday May 14, 2020 11:30am - 12:15pm
EW: Data Center Facility/Modular DC

11:30am

Panel: Network Orchestration and Automation
Large networks require network orchestration and automation software to manage the entire network lifecycle efficiently. This software almost always needs to support networks comprised of multiple network operating systems and multiple hardware device types.

In this panel, we will hear from commercial software providers as well as large-scale operators on how they’ve written software to support efficient network deployment and operations. In particular, they’ll share what they’d like to see in the underlying network operating systems (NOS) that run on the switches, as well as what the OCP networking ecosystem could focus on to make large-scale network orchestration/automation easier.

Moderators
avatar for Omar Baldonado

Omar Baldonado

Director of Software Engineering, Infra Network, Facebook
Omar Baldonado is the co-lead of the OCP networking project. At Facebook, he is part of the networking team, responsible for large-scale software systems that model, automate, monitor, and control Facebook’s network. The network domains include Facebook’s data centers; global... Read More →
avatar for Xin Liu

Xin Liu

Principal Product Manager, Microsoft Azure, Microsoft

Speakers
avatar for Alex Saroyan

Alex Saroyan

CEO, Netris
CEO of Netris - the new automatic netops platform that turns your physical network into service, and provides network services to your Kubernetes cluster, compute, and storage.
avatar for Sasha Ratkovic

Sasha Ratkovic

Founder & CTO, Apstra
Context Driven Automation
PS

Prince Sunny

Microsoft
avatar for Yanping Li

Yanping Li

Software Engineer, Facebook
Netsystems; Workflow Automation; Intent-based Network Management.


Thursday May 14, 2020 11:30am - 12:15pm
EW: Networking Software

11:40am

Can Block Storage Devices Converge on NVMe Protocol?
In the last ~40 years, some of us have seen many evolutions of block storage protocols come and go: IDE/ATA, SCSI, PATA, P-SCSI, SATA, USB, FCP, SAS … NVMe. Each protocol optimized specific Enterprise or Consumer storage device solutions and features. Regardless of the various storage media types and characteristics, is it finally possible for the storage industry to unify and consolidate around NVMe as the optimal block storage protocol? … Why? … When?

Speakers
MS

Matt Shumway

Office of the CTO, Seagate Technology
avatar for Mohamad H El-Batal

Mohamad H El-Batal

Office of the CTO, Seagate Technology
Mohamad El-Batal is the Seagate Enterprise Data Solutions (EDS) Chief Technologist, and part of the overall Seagate Office of the CTO team. He is given the opportunity to shape the Seagate EDS strategy and future storage product technology roadmap.In his career Mohamad lead a team... Read More →


Thursday May 14, 2020 11:40am - 12:10pm
EW: Storage Solutions

11:45am

Controllerless Open SDN Fabrics for Telco Edge Compute
This talk will dive into edge networking solutions using OCP platforms from EdgeCore with Pluribus Linux NOS with a focus on how the hardware and software architecture is optimized for small space and power constrained environemnts one one hand and simultaneously scaled edge deployments by leveraging the power of OCP hardware to integrate controllerless fabrics and multi-tenant routing on the other hand.

Furthermore, the talk will discuss how a fabric approach can provide deep network slicing to support multi-tenancy via segmentation in all three planes as a fundamental component of the architecture. The three planes - control, data and management - can be sliced to offer full isolation to avoid noisy neighbor syndrome.

The talk will also discuss how multiple federated fabrics will scale for multi-site data center deployments that will be found in a distributed low latency edge computing environment to provide new levels of automation and to avoid the complexity of provisioning heavy weight protocols like EVPN.

The talk includes OCP ready EdgeCore HW with Pluribus Linux NOS called out under: https://www.edge-core.com/productsKind.php?cls=1

Speakers
avatar for Chickayya Naik

Chickayya Naik

Senior Director of Engineering, Pluribus Networks
Chickayya Naik is a Senior Director of Engineering at Pluribus Networks leading a team of technical experts innovating in areas of Distributed Adaptive Cloud Fabric. He also drives architectural discussions with Tier-1 Telco Providers on NFVi, SDN, container networking and other topics... Read More →


Thursday May 14, 2020 11:45am - 12:00pm
EW: Telco & Edge

12:00pm

1U OCP Open Edge vRAN with Off-the-Shelf Acceleration
In this presentation, we study capabilities of the OCP Open Edge server for telecom far edge use cases. A case study for virtual RAN use is presented where OpenEdge is equipped with an off-the-shelf N3000 accelerator board supporting the OPAE open source framework. Studied edge DC solution demonstrates high physical and performance density. Flexibility of the solution to accelerating other edge NFV use cases is discovered as well.

Speakers
avatar for Mika Dementjeff

Mika Dementjeff

Project Manager, Nokia
avatar for Mikko Suni

Mikko Suni

Solution Architect, Intel


Thursday May 14, 2020 12:00pm - 12:15pm
EW: Telco & Edge

12:00pm

Open Rack V3 Universal Power Shelf Input Connector
As part of the Open Rack V3 project, we have been developing an open-sourced, multi-vendor solution to attach the power cord into the Open Rack V3 power shelf. This presentation will provide a detailed look at the goals, status, and specification for this connector.

Speakers
avatar for Will Stewart

Will Stewart

Data Center Segment Manager, Harting
avatar for Natesh Kannan

Natesh Kannan

Global Product Manager, Positronic
Natesh holds a M.S. in Smart Product Design from the Nanyang Technological University, Singapore. Natesh has 20 years of Engineering expertise with sound product innovation. Natesh is working with Positronic as a Global Product Manager for Power and Hybrid product line. Prior to this... Read More →


Thursday May 14, 2020 12:00pm - 12:30pm
EW: Rack & Power

12:00pm

LOG IN HERE for the Virtual Summit!
LOG IN HERE: https://onlinexperiences.com/scripts/Server.nxp?LASCmd=L:0&AI=1&InitialDisplay=1&ClientBrowser=0&ShowKey=86206

REGISTER HERE: https://www.opencompute.org/summit/global-summit/registration

Thursday May 14, 2020 12:00pm - 4:00pm

12:15pm

Enrich openEDGE Platform for 5G Solutions
Ever since Wiwynn joined Telco/openEDGE group, we actively participate the Eco-system and work with partners to offer more solutions. In this presentation, we will show an integrated solution based on Wiwynn EP100 with new networking sled designed by WNC, a leading communication solution provider. This in-chassis networking sled can act as an Ethernet switch without adding standalone switch and reduce the complexity within a limited space of cabinet.
As 5G communication is one of the most popular edge application, which is targeted by OCP openEDGE, Wiwynn and WNC will present the 5G RAN CU/DU as well as 5G core solutions based on Wiwynn EP100.

Speakers
GL

Gregary Liu

Director, Wiwynn
Server- and Rack-level design for the cloud services the customer would like to address!
avatar for Cheer Ko

Cheer Ko

Associate Vice President, WNC
Ethernet Switches for Enterprise, Datacenter and Edge


Thursday May 14, 2020 12:15pm - 12:30pm
EW: Telco & Edge

12:15pm

Managing Barbeques in Data Centers with Sustainability; Adaptability
“Density diversity” and “sustainability” are two top challenges/opportunities in 2020+ data center design. A multitude of liquid cooling solutions are emerging to address the challenges of density diversity – where needs for high density clusters are popping up in lower density data centers. Deployment of liquid solutions in operating data centers requires distribution strategies that are adaptable, “mission critical rated” and minimal risk on data center operations. Strategic evaluation of the elevated temperature cooling enabled by liquids expands sustainability options, both use of economizers as well as energy reuse and cooling optimization solutions.

Discussion will look at the categories of liquid based cooling solutions (row, rackdoor, chip, immersion) and strategies to deploy liquids that align with considerations of sustainability, adaptability, mission critical performance, and minimal risk on data center operations.

Speakers
avatar for Don Mitchell

Don Mitchell

Data Center Division Manager, Victaulic
Key topics: Liquid distribution & cooling solutions; density diversity; energy reuse; digital streamlining of data center constructionBackground - 20+ years in mission critical design, construction, operation, both data centers, nuclear submarines, missile systems
avatar for Tim Chambers

Tim Chambers

CEO, coolDC
Tim Chambers is Founder and CEO of coolDC. With over 30 years’ experience of working in the data centre and allied industries, he is no stranger to the challenges it faces.Tim has been vocal in his frustration concerning the industry’s reluctance to depart from the practice of... Read More →


Thursday May 14, 2020 12:15pm - 1:00pm
EW: Data Center Facility/Modular DC

12:30pm

Self-Building Data Center Networks
Growth in server capacity demand is triggering a high rate of new data center building turnup and network retrofits worldwide. Scaling out worldwide data center network build/retrofits requires a software defined approach to the whole planning, procurement, design, build execution process. Welcome to the world of “Self-Building” data center networks. In this talk we share our recent work in building a fully automated demand to bits pipeline to realize “self-building” data center networks. We will cover the key aspects of network build process automation from demand signal processing, topology and config staging, BOM generation as well as actual build/migration workflow automation.

Speakers
YC

Yan Cai

Software Engineer, Facebook Network Infra, Facebook
I am working on intent-based network design and management systems at Facebook.
SU

Sunil U. Khaunte

Software Engineering Director, Facebook


Thursday May 14, 2020 12:30pm - 1:00pm
EW: Networking Software

12:30pm

Survey of 48V-to-12V Architectures for On-Board DC/DC Conversion
With emerging use-cases driving the need for higher rack power and density, the adoption of higher bus voltage is becoming a popular choice to enable such server platforms. Higher bus voltages such as 48V is not a new concept within the rack power space as it is already widely used in telecom industry. Over the past few years, various distinct 48V architectures have been proposed and adopted to support on-board DC/DC conversion. This presentation is aimed to provide an overview of the existing 48V architectures and detailed survey of various conversion topology.

Speakers
avatar for Anthony Chan

Anthony Chan

Hardware Engineer, Facebook


Thursday May 14, 2020 12:30pm - 1:00pm
EW: Rack & Power

12:30pm

EDSFF: Dynamic Family of Form Factors for Data Center SSDs
Today all the EDSFF family of form factors share the same protocol (NVMe), the same interface (PCIe), the same edge connector (SFF-TA-1002), the same pinout and functions (SFF-TA-1009). There is a vast diversity of Enterprise and Datacenter applications. Data center infrastructure can be optimized in various ways for maximum capacity per rack unit, max performance, balanced performance to capacity, networking bandwidth, or for higher performance and TDP CPUs. Having a flexible and scalable family of form factors allows for optimization for different use cases, different media types on SSDs (e.g. TLC, QLC, SCM), scalable performance on PCIe 4.0 & 5.0, improving data center TCO through optimized power and thermal management, while maintaining key commonalities for compatibility and faster development. SFF-TA-1012 has been published to SNIA to show the breadth of the SFF-TA-1002 ecosystem, and provide pinout definitions for EDSFF SSDs (SFF-TA-1009), OCP NIC, Gen Z, PECFF, SNIA NVMe-oF ethernet drives, and future high speed devices.

Speakers
avatar for Jonmichael Hands

Jonmichael Hands

Sr. Strategic Planner, Intel
Strategic planner for Intel data center SSDsco-chair NVMe marketing workgroupco-chair SNIA SSSI SSD workgrouppresenting this year on EDSFF


Thursday May 14, 2020 12:30pm - 1:00pm
EW: Storage Solutions

12:30pm

Panel - Edge MEC Solutions, Tools and Orchestration
Wide ranging discussion of the challenges to optimize, deploy and lifecycle manage Edge Cloud deployments.

Moderator: Craig White

Speakers
avatar for Mike Moore

Mike Moore

Edge Cloud Platform Product Sales, OpenEdge Sub-Project Lead, Nokia
Over 20 years of R&D experience in both the Wireless and Defense Industries prior to transitioning into a customer facing sales and support role.  Now the Western Hemisphere Regional Product Manager for Nokia’s AirFrame Data Center Portfolio, with a focus on creating highly available... Read More →
avatar for Craig White

Craig White

Product Architect, Airframe R&D, Nokia
I have been in Open compute Telecom project from its creation. I was fortunate to have been Telecom project co-leader with Bill Carter. I work at Nokia in Mika Hattanpaa\\\'s Architecture R&D tribe lead by Samuli. I was responsible for Open Compute Foundation work and am currently... Read More →
EB

Elad Blatt

CSO, Silicom
avatar for Brandon Weilbacher

Brandon Weilbacher

Senior Director, Portfolio Management, Flex


Thursday May 14, 2020 12:30pm - 1:00pm
EW: Telco & Edge

1:00pm

Open Edge Deployment Options
This presentation highlights the main features of Open edge configurations for both indoor and outdoor installations. Examples of rack configurations, including power feed and networking options, are shown. This Nokia presentation outlines the configurations and deployment options spanning from a single chassis to configurations of multiple racks.

Speakers
avatar for Tomi Mannikko

Tomi Mannikko

Product Architect, Nokia


Thursday May 14, 2020 1:00pm - 1:20pm
EW: Telco & Edge

1:00pm

TMDC/TBLOCK - Tencent Modular Data Center On The Road
Introduce TMDC&TBLOCK technology largely deployed in Tencent infrastructure,and address backend logic to drive its definition and evolving path. Go through technical details over All-In-One micro-modular data center, including applicable environment, typical model of micro-module, power supply subsystem, cooling subsystem, structural subsystem, generic cabling subsystem, management and control subsystem, etc. Those subsystems and components in the micro-module can be prefabricated and qualified in the factory. Those components can be put into production after rapid onsite assembly and well scalability for dynamic volume adaptation.

Speakers
EL

Enso Li

Sr Datacenter Architect, Tencent


Thursday May 14, 2020 1:00pm - 1:30pm
EW: Data Center Facility/Modular DC

1:00pm

SONiC Telemetry In the Wild
In this session, we will share our development around SONiC telemetry supports, and our experiences of deploying telemetry services on SONiC in Alibaba's production data centers. The technical topics will cover SONiC gRPC telemetry implementation, streaming telemetry/real-time push notification, fine grain traffic monitoring and in-band telemetry etc.

Speakers
GW

Guohui Wang

Engineering Director, Alibaba


Thursday May 14, 2020 1:00pm - 1:30pm
EW: Networking Software

1:00pm

Silver BusBar Plating Connection Reliability Testing
An in-depth review of the busbar plating thickness requirements to show the possible benefit of using lubricants and to also confirm if the plating thickness can be reduced.  There will be multiple test methods used in this review which include insertion force test, insertion wear test, transportation test and fretting test.  Conclusions and results from each test will be shared with the community in this joint presentation.

Speakers
avatar for Caleb Lusk

Caleb Lusk

Engineering Systems Consultant, Rittal North America
avatar for Brian Kinkade

Brian Kinkade

Market Development Manager, Nye Lubricants
We have lubricants that have helped server OEMs resolve power failures due to the contact wear and corrosion at the power clip - bus bar interface. We are investigating how this applies to the OCP rack and power silver plated bus bar standard.


Thursday May 14, 2020 1:00pm - 1:30pm
EW: Rack & Power

1:00pm

Form Factors for the Future
EDSFF represents a new family of form factors. This talk will discuss the intersection of flash form factors, industry trends and hyperscale needs. In this talk Facebook seeks to share its experiences and insights into the challenges and solutions for the next generation of flash.

Speakers
avatar for Ross Stenfort

Ross Stenfort

Hardware System Engineer, Storage, Facebook
Ross Stenfort is a Hardware System Engineer at Facebook working on storage. He has been involved in development of SSDs, ROCs, HBAs and HDDs. He has over 40 granted patents. He has had extensive storage experience in both large and small companies including CNEX, Seagate, LSI, SandForce... Read More →


Thursday May 14, 2020 1:00pm - 1:30pm
EW: Storage Solutions

1:20pm

Lightning Talks: Intelligent Edge Use Cases
A discussion of AI, Analytics, Edge Acceleration and the Intelligent Edge Devices that will be required to deploy an optimizing Edge Cloud

Moderator: Sumithra Bhojan

Speakers
avatar for Brian Zahnstecher

Brian Zahnstecher

Principal, PowerRox
Brian Zahnstecher is a Sr. Member of the IEEE, Chair of the IEEE SF Bay Area Power Electronics Society (PELS), and the Principal of PowerRox, where he focuses on power design, integration, system applications, OEM market penetration, and private seminars for power electronics. He... Read More →
avatar for Dirk Blevins

Dirk Blevins

Telco Solution Architect, Intel
My name is Dirk Blevins. I am a Platform Architect at Intel and have been with the company for 18 years. My previous experience was in building printers for Lexmark Corporation and Military and Defense with Harris Corporation. I have done every job function imaginable in engineering... Read More →
YH

Yousef Hawwar

Edge Segment Lead, Intel
BW

Benjamin Wang

Edge Platform Architect, Foxconn


Thursday May 14, 2020 1:20pm - 2:05pm
EW: Telco & Edge

1:30pm

Goldstone - Open Source NOS for Optical Packet Transponder
In this presentation, I will introduce Goldstone - an open-source NOS for optical packet transponder.
It utilizes many existing open source components which have been developed in Open Compute Project (OCP) and Telecom Infra Project (TIP) including Open Network Linux (ONL), SONiC, Switch Abstraction Interface (SAI) and Transponder Abstraction Interface (TAI) to provide a full-fledged open source solution. ONL is used as the base operating system and provides a wide range of open network device support. On top of ONL, Kubernetes is employed to enable containerized application management, which realizes flexible and modular software composition. SONiC/SAI is deployed as a fleet of containers when the target hardware comprises Ethernet switch ASIC, whereas TAI is used when the target hardware has coherent transponder components. Because of its modular architecture, Goldstone can be extended to support networking devices, which don't have Ethernet ASIC, but may include conventional transponders, ROADMs or amplifiers in the future.

Goldstone was originally started as a prototype NOS for Edgecore's Cassini Platform by the proposal from NTT Electronics. This has led to a production deployment by mixi in Japan.

Speakers
avatar for Wataru Ishida

Wataru Ishida

Field Application Engineer, NTT Electronics America
avatar for Kingston Smiler Selvaraj

Kingston Smiler Selvaraj

CEO, Co-Founder, PalC Networks


Thursday May 14, 2020 1:30pm - 2:00pm
EW: Networking Software

1:30pm

Data Center System Stability Requirements for Power Supplies
Power supply units (PSU) are the most common type of rack load in a data center, and as such they must remain stable as the scale of the data center grows to several MW. In the past, Facebook has experienced data center power system instability as discussed at the Open Compute Project Data Center Facility Project Call (https://www.youtube.com/watch?v=aUk_-ufhR2Y&feature=youtu.be). The objective of this paper is to show the analytical process to establish system specifications for the PSU to remain stable within large scale data centers, which are characterized by having a large number of regulated power supply units as rack loads. The essential characteristic to derive system specifications is the impedance of the PSU. The paper will also give methods to calculate and measure the PSU impedance in practice.

Speakers
avatar for Mauricio Cespedes

Mauricio Cespedes

Power System Engineer, Facebook
avatar for Hamid Keyhani

Hamid Keyhani

Technical Lead, Facebook


Thursday May 14, 2020 1:30pm - 2:00pm
EW: Rack & Power

1:30pm

Optimizing Flash Form Factors for the Cloud
SSDs have gone from expensive storage devices used sparingly to a main-stream storage solution. In this transition, SSDs have left behind the legacy HDD defined form factors and are finding more optimal design opportunities including E1.S/L and E3.S/L. With E1.L SSDs, we can enable designs that can fit 1PB of flash in a front-serviceable system in the near future! On the other hand, E1.S SSDs can enable very high performance devices at a smaller capacity point that can be used across the storage and compute fleet. Together with the OCP community, we are working to ensure that an E1.S form factor is able to meet the needs of hyperscale consumers like us as well as the rest of the industry. We believe that these EDSFF formfactors are the future of flash, and will optimize and unlock new opportunities in our system designs.

Speakers
avatar for Jason Adrian

Jason Adrian

Storage Hardware Architect, Microsoft Azure
Jason is passionate about hyperscale and scale-out storage systems and storage architectures. His experience ranges from high-performance flash systems to archival systems, and everything in between.


Thursday May 14, 2020 1:30pm - 2:00pm
EW: Storage Solutions

2:00pm

FBOSS on SAI
Since its original announcement in 2014, FBOSS has scaled to power the Facebook network. However, the entire deployment uses the same family of switching ASICs and is built upon a shared ASIC programming codebase. This talk will present our experience porting a massively deployed NOS to run on top of SAI and the changes we made to our fast, robust testing suite to make that possible.

Speakers
JB

Jasmeet Bagga

Software Engineering Manager, Facebook
BB

Boris Burkov

Software Engineer, Facebook


Thursday May 14, 2020 2:00pm - 2:30pm
EW: Networking Software

2:00pm

Achieving Data Center Simplicity and Standardization with a Universal Rack PDU
There was a time when most data centers looked largely the same. No longer. Hyperscale facilities housing cloud and colocation providers bear little resemblance to today’s shrinking enterprise facilities, which increasingly act as a hub for hybrid networks stretching to the edge of the network. If that’s not enough, all of these facilities are evolving differently in ways both subtle and significant in different parts of the world.

The result is a data center ecosystem that is more complex than ever before. The Open Compute Project is, among other things, an effort to simplify that ecosystem, and the Vertiv Geist UPDU emerged from that work. The genesis for the development of the product was the overwhelming number of different rack power distribution units (PDUs) in use around the world. These systems all perform the same basic functions but with subtle and not-so-subtle differences and configurations that make inventory management a nightmare and procurement a maze. With a universal design, universal input and a detachable power cable, the Vertiv Geist UPDU is compatible with AC power topologies that are common worldwide.

This presentation will look at the challenges of complexity in today’s power topologies and the value of standardization through the prism of the Vertiv Geist UPDU. We will explore global nuances around power and power distribution and how the Vertiv Geist UPDU handles the many different power configurations.

Speakers
avatar for Brad Wilson

Brad Wilson

Vice President, Global Rack Power Distribution, Vertiv
Brad Wilson joined Vertiv in February 2018, as vice president of the global Rack Power Distribution Unit offering and engineering teams within Vertiv’s IT and Edge Infrastructure (ITEI) business.Prior to joining Vertiv, Brad was President of Geist, a privately held company specializing... Read More →
avatar for Grant Young

Grant Young

Offering Manager, Vertiv
As the Manager - Offering Management for the Global Rack PDU line of business at Vertiv, Grant Young is primarily responsible for developing and maintaining the global rPDU product offering as well as the rPDU engineer-to-order program. He utilizes voice of the customer, product data... Read More →


Thursday May 14, 2020 2:00pm - 2:30pm
EW: Rack & Power

2:00pm

Future Device Form Factors
With NVMe now becoming mainstream and new protocols such as CXL on the way system designers are challenged with supporting multiple device types in the front and rear of a server. This presentation discusses a new family of interoperable device form factors that support a wide range of device types through features such as a common high speed connector, multiple link widths, multiple PCB sizes, and a range of power profiles.

Speakers
avatar for Paul Kaler

Paul Kaler

Advanced Storage Technologist, HPE
WL

William Lynn

Server Advanced Engineering Architect, Dell


Thursday May 14, 2020 2:00pm - 2:30pm
EW: Storage Solutions

2:30pm

Intent-based Networking on SONiC
SONiC switches have been widely deployed in Alibaba data centers over the past few years. It has become the foundation of our rapid transition toward Intent-based Networking. In this talk, we will share our perspective on Intent-based Networking, explain why Intent-based Networking can significantly improve the stability and agility of our network infrastructure and demonstrate why SONiC is the most important enabler for us to adopt Intent-based Networking.

Speakers
GW

Guohui Wang

Engineering Director, Alibaba
avatar for Ennan Zai

Ennan Zai

Staff Engineer, Alibaba


Thursday May 14, 2020 2:30pm - 3:00pm
EW: Networking Software

2:30pm

SSD Thermal Simulation Standardization
The goal of establishing a standardized thermal analysis approach for SSD devices is to align both producers and consumers in the area of form factor selection, thermal analysis, device comparison, and spatial boundaries. It is a known problem that thermal designers and technicians have a variety of methodologies by which they qualify their deviceís designs. The outcome of this effort will combine best known methodologies for SSD characterization into one industry specification and approach that establishes confidence in comparing key engineering metrics

Speakers
DL

Dave Landsman

Dir. of Industry Standards, Western Digital
PG

Paul Gwin

Engineer, Intel


Thursday May 14, 2020 2:30pm - 3:00pm
EW: Storage Solutions

3:00pm

Modular Chassis, MPLS, and MACsec Integration on SONiC
In order to be truly ubiquitous throughout the data center and WAN core, SONiC must embrace new features, platform models, and deployment cases. We demonstrate MPLS and MACsec with MKA on Juniper’s newest 400G modular chassis, the PTX10008. Further, we discuss how hot swappable devices (FRU insertion and removal in a modular chassis) affect SONiC dynamic port management.

Speakers
avatar for Lester Bird

Lester Bird

Distinguished Engineer, Juniper Networks
Lester has worked for 33 years in computer networking. He specializes in network stacks and distributed, fault tolerant systems. He is currently leading the multi-PFE SONiC project at Juniper Networks.


Thursday May 14, 2020 3:00pm - 3:30pm
EW: Networking Software

3:00pm

Enabling Robust JBOD designs Via SPL Design Targets
The objective of this presentation is to share the efforts Facebook Storage team has made on how to quantitively gauge the system acoustic vibration impact on the future HDDs and establish a reference acoustic vibration guideline for the next generation HDD storage platform.
Since the number of drives is fixed after the chassis design is done, the main driving force for the HDD storage platform at Facebook is to increase the drive density and achieve the large TCO win. However, higher capacity drives are more sensitive to vibration, most impact coming from acoustic vibration.
The current approach relies on the readiness of the actual drives from the suppliers to evaluate acoustic vibration impact, is limited to current-gen drives and is lacking in insights about the platform compatibility with the future drives. Although emulators are available from some suppliers for the purpose of evaluating this, the methods today are proprietary sometimes and vendor specific.
Sound Pressure Level (SPL) around the hard drives is used in this study as the indicator for the acoustic vibration impact. SPL is a drive agnostic characteristic and only depends on the system fan and chassis design. It would be really beneficial to the industry if all the HDD vendors can publish SPL guidelines to stay under (preferably a mask of sorts, that is constant across multiple HDD generations), for the purpose of acoustic vibration mitigation in a chassis from its early product design phase.

Speakers
avatar for Madhavan Ravi

Madhavan Ravi

Storage Hardware Engineer, Facebook
Madhavan is a Hardware engineer in the Storage Hardware team that designs all the Storage gear used inside Facebook. His interests include hardware architecture, storage system efficiency and performance, storage media, and file systems.
avatar for Jun Shen

Jun Shen

Thermal Engineer, Facebook
Jun is a thermal engineer in the Infra HW group. His work at Facebook is primarily in the storage platform thermal management. He earned his PhD degree in Mechanical Engineering from the University of Maryland, College Park in 2004.


Thursday May 14, 2020 3:00pm - 3:30pm
EW: Storage Solutions
 
Friday, May 15
 

8:00am

LOG IN HERE for the Virtual Summit!
LOG IN HERE: https://onlinexperiences.com/scripts/Server.nxp?LASCmd=L:0&AI=1&InitialDisplay=1&ClientBrowser=0&ShowKey=86206

REGISTER HERE: https://www.opencompute.org/summit/global-summit/registration

Friday May 15, 2020 8:00am - 4:30pm

9:00am

OCP Future Technologies Symposium Kick-Off
Speakers
avatar for Bill Carter

Bill Carter

CTO, OCP
Bill is CTO for the Open Compute Project Foundation where he is a visionary and pathfinder on this open source journey. Bill works with the entire OCP Community and many other software groups, standards organizations, and industry consortiums to bring together people from around the... Read More →
avatar for Lesya Dymyd

Lesya Dymyd

Innovation Engineer, 2CRSI
Primary lead for the successful execution of an event, including communication with academic institutions and leading the Program Committee. Lesya holds a doctorate degree from the University of Strasbourg and was a visiting researcher at ETH Zurich. She is currently a strategic innovation... Read More →


Friday May 15, 2020 9:00am - 9:05am
OCP Symposium

9:00am

OCP Experience Center
OCP Accelerator Module, OAI ( Facebook, Hyve Design Solutions, Inspur, Xilinx)
OAI will demonstrate currently available OAMs as well as show functional implementations in multiple OAI systems.

OpenEDGE (Artesyn Embedded Power, ASUS, BelPower Solutions & Protection, Flex, Inventus Power, MiTAC, NOKIA)
OCP Open Edge project creates open specifications, standards and reference designs for Edge Computing and Telco use cases that require small form factor server designs that enable high performance and low latency computing. Participants of the project are across the ecosystem including Telco Service Providers, OCP solution providers, ODM/OEM manufacturers, and key technology providers. Nokia has contributed the initial Open Edge solution that creates the basis for interoperable designs.

RunBMC (Dropbox, Intel, Vertiv)
This is a demo of hardware modules following OCP RunBMC specification. The first video delivered by authors of the specification from Dropbox features the demo of HS2500 Prototype Module and HS2500 Production Module. This video also presents the standalone RunBMC Bring Up Vehicle and RunBMC-compatible Olympus chassis with Nuvoton Poleg RunBMC Module. The second video delivered by Intel presents FPGA-based RunBMC module featuring Intel® Cyclone V and Intel® MAX10. The video demonstrates the integration of this module in standard Intel 2S Skylake-X platform by using RunBMC Bring Up Vehicle card from Dropbox. In this video you can also see OpenBMC boot on Intel RunBMC module and basic management of the platform with WebUI, SOL and Remote KVM.

Open System Firmware, OSF (Facebook, Wiwynn, HP Enterprise, Intel and ITRenew).
Open system firmware is an open development project, the goal of which is to allow OCP owners to 'own their firmware' -- to move the point of control of firmware to the system owner. Starting in March, 2021, OCP badging for servers will require that systems support OSF. We are exploring multiple ways to accomplish this with components including Hostboot, UEFI, coreboot, and LinuxBoot.

OpenRMC (Inspur)
The Demo will show the establishment of OpenRMC Northbound 1.0 profile execution on top of the bundle of OpenRMC FW & HW (RMC) contributed to OCP as a complete reference solution to address the fulfilment of user's requirements/demand for a comprehensive management of an entire rack system and beyond via RESTful API through remote management portal/client console with detailed WebUI and paving the way for exploring future OpenRMC v2.0.

SEBA Solution (MITAC and EdgeCore)
SEBA solution is one of ONF project and with Edgecore and MiTAC, you can have the OCP hardware ready for SEBA.

OCP NIC 3.0 group (Amphenol, Broadcom, Dell Technologies, Facebook, Intel, Marvell Semiconductor, NVIDIA Networking - Mellanox, Quanta, TE Connectivity, UNH-IOL, Wiwynn Corporation)
The goal of this subgroup is to form a healthy OCP NIC 3.0 ecosystem through a supportive, informative and inclusive community.

96Tb DDC based Disaggregated Router ( Credo, DriveNets, Edgecore Networks, Keysight Technologies, UfiSpace )
The Open Compute Project (OCP) has selected UfiSpace, Credo and DriveNets to give the first ever live demonstration of a Distributed, Disaggregated Chassis (DDC) model as proposed to the OCP by ATT. This demonstration is the largest live demonstration of a distributed router cluster ever given in public. The technology is aligned with the vision of TIP (Telecom Infra Project) who also invited DriveNets and UfiSpace to advise on their Disaggregated Open Router Initiative and ensure its alignment with the OCP's efforts around network disaggregation.

Friday May 15, 2020 9:00am - 4:00pm
Experience Center

9:00am

Expo Hall Open
Friday May 15, 2020 9:00am - 4:00pm
Expo Hall

9:00am

9:05am

OCP Symposium Direction and Heading
OCP Chair lays out two-year strategic vision around adoption and something thoughts on how we might get there, and some anticipated challenges.

Speakers
avatar for Allan Smith

Allan Smith

Area 404 Lab Manager, Facebook
Allan is a graduate of the US Air Force Academy and ESC Toulouse. He is currently the Lab Manager for Area 404, Facebook’s hardware prototyping laboratory. Prior to joining Facebook, Allan led Rapid Response Space Works efforts with NASA Ames Research Center.


Friday May 15, 2020 9:05am - 9:25am
OCP Symposium

9:25am

The Importance of The Future Technologies Symposium for OCP
The talk is divided in three portions. The first part of the talk will deal with motivation of OCP and bit of a historic perspective of OCP. Second part of the talk will address some of the challenges ahead of the OCP community and the infrastructure development in light of growing workloads driven by ML/AI and other applications raising exponentially. This sets the stage for value of Symposium and how it can tie into OCP in the coming decade. Specific example areas of engagement will be proposed along with vision for where OCP is focusing. The theme will tie with an ending that proposes how Symposium can add additional value for growing needs of OCP community.

Speakers
avatar for Dharmesh Jani

Dharmesh Jani

OCP IC Co-Chair and FB Open Ecosystem Lead, Facebook
Dharmesh Jani (‘DJ’) has been an active member of OCP since 2012. DJ has over 20+ years of experience in various roles spanning engineering, product management, and business strategy. He started his career at Rockwell Science Center designing ultra-high speed circuits in CMOS... Read More →


Friday May 15, 2020 9:25am - 9:40am
OCP Symposium

9:40am

Delivering the OpenEdge: New opportunities for collaboration
With the demand for low latency and large data set applications expected to grow substantially,
compute is being distributed to many edge locations which may be operated by cloud service providers,
enterprises or communications service providers. These edge computing deployments might be a mini-
data center installed in a telco central office, several racks of enterprise equipment dropped on a factory
floor, a micro DC contained in a wall box in a retail outlet, or an autonomous vehicle. There are many
new challenges with moving compute to the edge that will benefit from tight university and industry
collaboration solve. These challenges include new innovative designs allowing user customizations at
new price points for hardware, software stacks that allow developers to write once and deploy in many
locations, and management tools enabling IT administrators to manage 100’s to 1,000s of geographically
distributed computing sites. This keynote will examine the role that open infrastructure, hardware and
software, is expected to play and areas for collaboration in the multi-year edge computing journey
enabling innovation, security, customization and cost reduction.

Speakers
avatar for Cliff Grossner

Cliff Grossner

Executive Director Research & Technology Fellow, Omdia (formerly IHS Markit)
\"As head of the Cloud and Data Center Research Practice, he is responsible for research quality across the practice and his research is focused on cloud services, data center compute and networking, and software defined networking (SDN). Dr. Grossner has more than 25 years of telecommunications... Read More →


Friday May 15, 2020 9:40am - 10:00am
OCP Symposium

10:00am

Academic Perspective on OCP Symposium
This talk discusses the connection between academia and OCP from the academia perspective. We will highlight the need of OCP to attract more academia participants, and the benefits for academia to engage
with OCP. A few opportunities to enhance the academia presence in OCP will be discussed, followed by an open discussion on future actions towards this direction.

Speakers
avatar for Minjie Chen

Minjie Chen

Assistant Professor of Electrical Engineering, Princeton University


Friday May 15, 2020 10:00am - 10:15am
OCP Symposium

10:15am

Industry Investment Panel
The industry investment panel is the discussion with the experts from the leading organizations on the questions of investments, protection of IP rights, research and development of new IT and data center solutions within the existing ecosystems. The panelists discuss the importance of collaboration between the universities, startups and member organizations of the global Open Compute Project community and give several potential technological areas for future collaborative projects.

Moderators
avatar for Lesya Dymyd

Lesya Dymyd

Innovation Engineer, 2CRSI
Primary lead for the successful execution of an event, including communication with academic institutions and leading the Program Committee. Lesya holds a doctorate degree from the University of Strasbourg and was a visiting researcher at ETH Zurich. She is currently a strategic innovation... Read More →

Speakers
avatar for Rick Calle

Rick Calle

Emerging Technologies BD, M12 (Microsoft)
avatar for Uri Cummings

Uri Cummings

CTO, Intel
avatar for James Stewart

James Stewart

Director, Optical Hardware Development, Facebook


Friday May 15, 2020 10:15am - 11:15am
OCP Symposium

11:15am

11:30am

Container-based NFVI Platform with Network Performance Analysis and Tuning for Mobile Edge Private Network
Speakers
avatar for Wei-En Liang

Wei-En Liang

Software Engineer, Industrial Technology Research Institute
Software / Infra. Engineer @ ITRIemail: jace.liang@itri.org.tw


Friday May 15, 2020 11:30am - 11:45am
OCP Symposium

12:00pm

Toward Intelligent Interconnect Fabrics
Speakers
avatar for Tony Chan Carusone

Tony Chan Carusone

Department of Electrical and Computer Engineering, University of Toronto
My lab conducts research on integrated circuits for digital communication.


Friday May 15, 2020 12:00pm - 12:15pm
OCP Symposium

12:00pm

LOG IN HERE for the Virtual Summit!
LOG IN HERE: https://onlinexperiences.com/scripts/Server.nxp?LASCmd=L:0&AI=1&InitialDisplay=1&ClientBrowser=0&ShowKey=86206

REGISTER HERE: https://www.opencompute.org/summit/global-summit/registration

Friday May 15, 2020 12:00pm - 4:00pm

12:15pm

Programmable integrated optical circuits using the open-access CORNERSTONE platform
Speakers
avatar for Callum Littlejohns

Callum Littlejohns

Principal Enterprise Fellow, University of Southampton
I am the coordinator of an open access silicon photonics foundry service for rapid prototyping of devices and circuits.


Friday May 15, 2020 12:15pm - 12:30pm
OCP Symposium

12:30pm

2.5D Heterogenous Silicon Photonics Light Engine with Integrated DFB Lasers and Electronics
Speakers
avatar for Radhakrishnan Nagarajan

Radhakrishnan Nagarajan

CTO & SVP, Inphi Corp
Radha Nagarajan is the CTO and SVP, at Inphi. Prior to joining Inphi, he was with Infinera, as a Fellow.He is a Fellow of the IEEE, OSA and IET. In 2006, he shared the IEEE/LEOS Aron Kressel Award for his work on commercializing large-scale photonic integrated circuits. He has a Ph.D... Read More →


Friday May 15, 2020 12:30pm - 12:45pm
OCP Symposium

12:45pm

Towards peta-bit/s optical links
Speakers
FA

Firooz Aflatouni

Univ. of Pennsylvania
FA

Farshid Ashtiani

Univ. of Pennsylvania


Friday May 15, 2020 12:45pm - 1:00pm
OCP Symposium

1:00pm

Coherent Co-Packaged Optical Interfaces Enabling Next-Generation Data Center Switching
Speakers
avatar for Brandon Buscaino

Brandon Buscaino

Facebook
I am a PhD student at Stanford University working under Prof. Joseph M. Kahn in the optical communications research group. I primarily research novel architectures for data center co-packaged optics including designs using external and integrated light sources. I also work on resonator-enhanced... Read More →


Friday May 15, 2020 1:00pm - 1:15pm
OCP Symposium

1:15pm

Using Faucet to Validate Hardware OpenFlow Implementations
Speakers
avatar for Brad Cowie

Brad Cowie

Research Programmer, University of Waikato
Brad is a member of the WAND Network Research Group at the University of Waikato. He is also a core member of the FAUCET project which develops an open source OpenFlow controller for enterprise networks. Utilising his years of experience deploying servers and networks he does SDN... Read More →


Friday May 15, 2020 1:15pm - 1:30pm
OCP Symposium

1:30pm

Towards AI-assisted Video Compression
Speakers
avatar for Sicheng Li

Sicheng Li

Research Scientist, Alibaba Group


Friday May 15, 2020 1:30pm - 1:45pm
OCP Symposium

1:45pm

Distributed Shared Memory Architecture
Speakers
DX

Di Xu

Alibaba Group


Friday May 15, 2020 1:45pm - 2:00pm
OCP Symposium

2:00pm

ReFlex4ARM: Supporting 100GbE Flash Storage Disaggregation on ARM SoC
Speakers
avatar for Minghao Xie

Minghao Xie

PhD Student, UC Santa Cruz
I am a second-year computer engineering Ph.D. student at Baskin School of Engineering, UC Santa Cruz. I am fortunate to be advised by Chen Qian and Heiner Litz. My research interests revolve around computer networks and systems. My current work focuses on flash storage aggregation... Read More →


Friday May 15, 2020 2:00pm - 2:15pm
OCP Symposium

2:15pm

2:30pm

Hybrid and Resonant Switched-Capacitor Converters: Achieving Ultra Efficient and Compact High Step-Down DC-DC Power Conversion for Datacenter Applications
Speakers
avatar for Zichao Ye

Zichao Ye

UC Berkeley
Hi there, I am a last year PhD student studying power electronics in the EECS department of UC Berkeley. My research focus is on hybrid switched-capacitor power converters. One of its emerging application is ultra-efficient and compact power delivery fot future data centers. I am... Read More →
WC

Wen Chuen Liu

UC Berkeley


Friday May 15, 2020 2:30pm - 2:45pm
OCP Symposium

2:45pm

LEGO-PoL: A Family of Extreme Performance Miniaturized Point-of-Load Converters for High Current Microprocessors
Speakers
avatar for Jaeil Baek

Jaeil Baek

Postdoctoral Researcher, Princeton University
Jaeil Baek is a Postdoctoral Research Associate of Electrical Engineering of Princeton University. He received a Ph.D. from the Korea Advanced Institute of Science and Technology in 2018. His research interests include point-of-load converter, grid-interface power electronics, digital... Read More →
YE

Youssef Elasser

Princeton University


Friday May 15, 2020 2:45pm - 3:00pm
OCP Symposium

3:00pm

3:15pm

Wide Voltage Range High-Efficiency Sigma Converter 48V VRM With Fast Transient Response
Speakers
avatar for Xin Lou

Xin Lou

Virginia Polytechnic Institute and State University


Friday May 15, 2020 3:15pm - 3:30pm
OCP Symposium

3:30pm

Zoned Namespaces: From Archival to I/O Predictability
Speakers
avatar for Javier Gonzalez

Javier Gonzalez

Pricipal Software Engineer, Samsung
Javier is a Principal Software Engineer at Samsung Electronics, where he founded and leads SSDR (Samsung Electronics Denmark Research). He is a contributor to the Linux kernel and other open-source projects. His expertise is in building new ecosystems for upcoming technologies. He... Read More →


Friday May 15, 2020 3:30pm - 3:45pm
OCP Symposium

3:45pm

Reconfigurable Array of Inexpensive Batteries Architecture
Speakers
avatar for Tzi-cker Chiueh

Tzi-cker Chiueh

Vice President & General Director, Industrial Technology Research Institute
Dr. Chiueh is the Vice President and General Director of Information and Communications Research Laboratories and was a General Director of Cloud Computing Center for Mobile Applications of Industrial Technology Research Institute, and a Professor of Computer Science Department, Stony... Read More →


Friday May 15, 2020 3:45pm - 4:00pm
OCP Symposium

4:00pm

OCP Future Technologies Symposium Winner Announcement & Wrap-Up
Speakers
avatar for Dirk Van Slyke

Dirk Van Slyke

Director of Marketing, Director, Marketing & Events
Dirk began his career over thirty years ago working in brand management leadership for full-service, creatively-driven branding agencies such as The Richards Group and GSD&M. He has led successful marketing initiatives in a variety of industries, including ecommerce, healthcare, fashion... Read More →
avatar for Lesya Dymyd

Lesya Dymyd

Innovation Engineer, 2CRSI
Primary lead for the successful execution of an event, including communication with academic institutions and leading the Program Committee. Lesya holds a doctorate degree from the University of Strasbourg and was a visiting researcher at ETH Zurich. She is currently a strategic innovation... Read More →
avatar for Allan Smith

Allan Smith

Area 404 Lab Manager, Facebook
Allan is a graduate of the US Air Force Academy and ESC Toulouse. He is currently the Lab Manager for Area 404, Facebook’s hardware prototyping laboratory. Prior to joining Facebook, Allan led Rapid Response Space Works efforts with NASA Ames Research Center.


Friday May 15, 2020 4:00pm - 4:05pm
OCP Symposium