NVIDIA at OCP Summit 2024
San Jose McEnery Convention Center
October 15–17
This year, NVIDIA is participating in the Open Compute Global Summit (OCP) at the San Jose McEnery Convention Center from October 15–17 to showcase the latest advancements and innovations in NVIDIA’s accelerated computing platform. We’ll feature a keynote and groundbreaking sessions from NVIDIA speakers on the latest in AI, networking, energy efficiency, security, and more.
NVIDIA Contributes Blackwell Platform Design to Open Hardware Ecosystem, Accelerating AI Infrastructure Innovation
At this year’s OCP Global Summit, NVIDIA will be sharing key portions of the NVIDIA GB200 NVL72 system’s electro-mechanical design with the OCP community — including the rack architecture, compute and switch tray mechanicals, liquid-cooling and thermal environment specifications, and NVIDIA NVLink™ cable cartridge volumetrics—to support higher compute density and networking bandwidth.
Schedule at a Glance
Explore a wide range of groundbreaking sessions in the field of AI, accelerated computing, networking, and more. Take a closer look at the scheduled NVIDIA sessions that are part of this year’s program.
| Tuesday 10/15 | Wednesday 10/16 | Thursday 10/17 |
| Keynote: Fostering Collaboration: Designing Data Centers for Tomorrow's AI Workloads 9:24–9:36 a.m. PT Grand Ballroom 220 |
The Exponential Demands AI Places on the Rack and Data Center 8:05–8:25 a.m. PT Concourse Level - 220B |
SIOV R2: Next-Gen I/O Virtualization Framework for Next-Gen Data Centers 9:10–9:30 a.m. PT Concourse Level - 210CG |
| Empowering AI Networking With SONiC 1:10–1:30 p.m. PT Lower Level - LL20BC |
OCP Data Formats for Deep Learning 8:25–8:45 a.m. PT Concourse Level - 220B |
OCP Cooling Environments and ASHRAE TC9.9: Roadmap for Future Collaboration 9:45–10:15 a.m. PT Concourse Level - 210AE |
| Exploring the Wilderness: Optimizing Ethernet Fabrics for AI Workloads 1:30–1:50 p.m. PT Lower Level - LL20BC |
Standardizing Hyperscaler Requirements for Accelerators 9:00–9:30 a.m. PT Concourse Level - 2210CG |
Chiplet Interoperability and Breakthroughs in AI 10:45–11:05 a.m. PT Concourse Level - 220B |
| Mature at Scale Memory Fabrics for all Performance and Price Points 1:40–2:00 p.m. PT Lower Level - LL20A |
Fabric Resiliency at Scale 10:25–10:45 a.m. PT Lower Level - LL20A |
NVIDIA MGX: Faster Time to Market for the Accelerated Data Center 10:50–11:10 a.m. PT Concourse Level - 210CG |
| Building the First SONiC Cloud AI-Benchmarked Cluster in the World 2:50–3:10 p.m. PT Lower Level - LL20BC |
New Approaches to Network Telemetry: Essential for AI performance 10:45–11:00 a.m. PT Lower Level - LL20A |
Roadmap for a Durable Chip Coolant Temperature 12:30–12:50 p.m. PT Concourse Level - 210AE |
| Deploying the AI Factory 10:45–11:00 a.m. PT Lower Level - LL20D |
Cold Plate Panel Discussion on DLC for AI TTM 2:00–3:00 p.m. PT Lower Level - LL20BC |
|
| Compute Resilience: Industry Update 11:00–11:30 a.m. PT Concourse Level - 2210CG |
Study of Fluid Velocity Limits in Cold Plate Liquid Cooling Loops 3:20–3:40 p.m. PT Concourse Level - 210AE |
|
| High Performance Data Center Storage Using DPUs 1:10–1:30 a.m. PT Concourse Level - 210DH |
||
| AI for Quantum Computing and NVIDIA’s QC Strategy 1:10–1:30 a.m. PT Lower Level - LL20BC |
||
| Introducing the SPDM Authorization Specification 1:30–1:50 p.m. PT Concourse Level - 220C |
||
| Caliptra Journey: What Are We Doing Next? 2:10–2:30 p.m. PT Concourse Level - 220C |
||
| CTAM:Compliance Tool for Accelerator Manageability 4:05–4:25 p.m. PT Concourse Level - 210CG |
||
| OCP Streaming Boot Implementation Update 4:10–4:30 p.m. PT Concourse Level - 220C |
||
| Protecting AI Workloads From Noisy Neighbors in Cloud Networks With NVIDIA Spectrum-X 4:10–4:30 p.m. PT Concourse Level - 220B |
||
| Towards an Open System for AI 4:30–5:00 p.m. PT Concourse Level - 22S |
Architectures for the Modern Data Center
Explore NVIDIA Solutions
Programs and Technical Training
Meet Our Partners
Read this blog on NVIDIA’s contribution of NVIDIA GB200 NVL72 designs to OCP.