# NVIDIA at OCP Global Summit 2025 San Jose McEnery Convention Center October 13–16 Check out this year’s Open Compute Global Summit (OCP) to see all the latest advancements in the NVIDIA accelerated computing platform. We’ll feature a keynote, plus groundbreaking sessions from NVIDIA speakers on the latest in AI, AI infrastructure, networking, energy efficiency, security, and more. ![OCP25 NVIDIA Keynote](https://www.nvidia.com/content/dam/en-zz/Solutions/events/ocp-summit/2025/open-infrastructure-for-ai.jpg) Consent for Optional Cookies (googleCookiePolicyLink)YouTube sets performance, advertising, and other optional cookies(/googleCookiePolicyLink) when you watch embedded videos. To watch this video, you need to turn on optional cookies for the site. By clicking “Accept and Play Video,” you will automatically turn on advertising and other optional cookies for the site and accept our (nvidiaTermsOfServiceLink)Terms of Service(/nvidiaTermsOfServiceLink) (which contains important waivers). Please see our (nvidiaPrivacyPolicyLink)Privacy Policy(/nvidiaPrivacyPolicyLink) and (nvidiaCookiePolicyLink)Cookie Policy(/nvidiaCookiePolicyLink) for more information. Cancel Accept and Play Video Alternatively, you can (youtubeLink)watch this video on YouTube(/youtubeLink). ## OCP25 NVIDIA Keynote Watch Ian Buck, VP of HPC, explain how the NVIDIA ecosystem and rack architectures are powering the most demanding AI workloads. --- ## News at OCP ### NVIDIA Partners Drive Next-Gen Open, Efficient Gigawatt AI Factories Partners are tapping NVIDIA Vera Rubin NVL144, NVIDIA MGX™, and NVIDIA Kyber for energy-efficient AI factories. [Read the Blog](https://blogs.nvidia.com/blog/gigawatt-ai-factories-ocp-vera-rubin) ### NVIDIA Spectrum-X™ Ethernet Switches Speed Up Networks for Meta and Oracle Hyperscalers adopt NVIDIA open networking to drive gigascale AI data center performance. [Read the Press Release](https://nvidianews.nvidia.com/news/nvidia-spectrum-x-ethernet-switches-speed-up-networks-for-meta-and-oracle) ### Building the 800 VDC Ecosystem for Efficient, Scalable AI Factories Generative AI has transformed traditional data centers into power-focused “AI factories,” requiring a complete redesign centered on efficiency, scalability, and high energy demands. [Read the Blog](https://developer.nvidia.com/blog/building-the-800-vdc-ecosystem-for-efficient-scalable-ai-factories/) ## Schedule at a Glance OCP 2025 spotlights the latest innovations driving the future of open data center design and accelerated infrastructure. Explore the powerful lineup of NVIDIA sessions featured in this year’s program to see how we’re shaping the next era of scalable, sustainable computing. | | | | | | --- | --- | --- | --- | | Monday 10/13 | Tuesday 10/14 | Wednesday 10/15 | Thursday 10/16 | | | | Revolutionizing Networking for the Era of Agentic AI 9:15–9:30 a.m. Concourse Level, 210CDGH | Technical Paths to the New Era of GPU-Initiated Storage 8:40–8:55 a.m Concourse Level, 220C | | | | Introducing...UQDv2.0! 9:45–10:00 a.m. Concourse Level, 211 | Device Ownership Transfer and Cryptographic Binding for Device Without Secure Storage 9:10–9:30 a.m. Concourse Level, 211 | | | | Methods to Support Higher-Power Rack Cooling 10:25–10:40 a.m. Concourse Level, 211 | OCP Silent Data Corruption Workgroup Update 9:10–9:30 a.m. Concourse Level, 212 | | | Cooling Environments Panel 1:00–1:30 p.m. Lower Level, LL20D | Panel: Progress in Optical Interconnects for AI Clusters 11:05–11:30 a.m. Concourse Level, 210CDGH | Climate-Based Optimization of Cooling Strategies for Sustainable Data Centers 10:05–10:20 a.m. Lower Level, LL20D | | | Guidelines for Pre-Commission Preparation of Technology Cooling System (TCS) Row Manifolds in Liquid-Cooled Data Centers Presentation 1:35–1:50 p.m. Lower Level, LL20D | 800 VDC MGX Accelerated Computing Rack & Energy Storage for Improved GPU Power Utilization 11:10–11:30 a.m. Lower Level, LL20D | Panel: Standardizing GPU Management: Redfish, Telemetry, and Firmware Update Protocols 12:30–1:00 p.m. Concourse Level, 212 | | | OCP Cooling Environments—ASHRAE 2:15–2:45 p.m. Lower Level, LL20D | Data Center Facility Project Highlights 12:30–12:45 p.m. Concourse Level, 210CDGH | SONiC for AI Networking: Looking Back and Ahead 1:00–1:25 p.m. Concourse Level, 230A | | | Empowering AI at Scale With Lenovo Neptune® Liquid Cooling 2:45–3:10 p.m. Concourse Level, 210ABEF | Operating Temperatures to Support Performance, Energy Efficiency, and Heat Recovery 1:10–1:25 p.m. Concourse Level, 210CDGH | Revisiting Jim Gray’s 5-Minute Rule in the Storage Next Era: A First-Principles Perspective 1:55–2:10 p.m. Concourse Level, 210ABEF | | | The Value of Digital Twins: Real-World Case Studies From the OCP Community 3:35–3:50 p.m. Lower Level, LL21B | Rack Emulator Design for Multigenerational IT Racks 3:50–4:05 p.m. Concourse Level, 210CDGH | DC-Stack: Challenges and Lessons Learned 3:45–4:00 p.m. Lower Level, LL20BC | | Ian Buck Keynote: Shaping the Future of Open Infrastructure for AI 4:35–4:50 p.m. Street Level - South Hall | Using MGX Racks and Digital Twins to Convert Any Data Center Into an AI Factory 3:55–4:10 p.m. Lower Level, LL21B | Panel: Dirty Insights Into Liquid Cooling—Cleaning & Commissioning the TCS 4:30–5:00 p.m. Concourse Level, 210CDGH | | ## Architectures for the Modern Data Center ##### NVIDIA Blackwell GPU Architecture Explore the groundbreaking advancements the NVIDIA Blackwell architecture brings to generative AI and accelerated computing. Building upon generations of NVIDIA technologies, Blackwell defines the next chapter in generative AI with unparalleled performance, efficiency, and scale. [Learn More About Blackwell](https://www.nvidia.com/en-us/data-center/technologies/blackwell-architecture.md) ##### NVIDIA Grace CPU Architecture The NVIDIA Grace™ CPU delivers exceptional performance, power efficiency, and high-bandwidth connectivity for HPC and AI applications. The NVIDIA Grace Hopper™ Superchip is a breakthrough integrated CPU+GPU for giant-scale AI and HPC applications. For CPU-only HPC applications, the NVIDIA Grace CPU Superchip provides superior performance, memory bandwidth, and energy efficiency. [Learn More About Grace](https://www.nvidia.com/en-us/data-center/grace-cpu.md) ##### NVIDIA MGX Architecture NVIDIA MGX™ is a modular reference architecture for accelerated computing that supports hundreds of GPU, DPU, CPU, storage, and networking combinations for AI, high-performance computing (HPC), and NVIDIA Omniverse™ workloads. [Learn More About MGX](https://www.nvidia.com/en-us/data-center/products/mgx.md) ##### NVIDIA Spectrum-X Ethernet Network Platform NVIDIA Spectrum-X™ Ethernet delivers the highest performance for AI. It connects compute fabrics within the data center and scales across multiple AI data centers with Spectrum-XGS Ethernet technology to form massive AI super-factories capable of giga-scale intelligence. [Learn More About Spectrum-X](https://www.nvidia.com/en-us/networking/spectrumx.md) --- ## Explore NVIDIA Solutions ### Think SMART. Think NVIDIA Inference. As reasoning models generate exponentially more AI tokens, demand for compute surges. Full-stack inference optimization is the key to ensuring that you're thinking smart about scaling AI at AI factory scale. [Learn More About AI Inference](https://www.nvidia.com/en-us/solutions/ai/inference.md) ### NVIDIA AI Factories Unlike traditional data centers, an AI factory is specifically designed to manufacture intelligence at scale and, with specialized AI infrastructure, support the intensive computational needs of AI workloads. These factories excel at AI reasoning, [agentic AI](https://blogs.nvidia.com/blog/what-is-agentic-ai/), and [physical AI](https://www.nvidia.com/en-us/glossary/generative-physical-ai.md), enabling faster, more accurate decision-making across industries. [Learn More About AI Factories](https://www.nvidia.com/en-us/solutions/ai-factories.md) ### NVIDIA GB300 NVL72 The NVIDIA GB300 NVL72 features a fully liquid-cooled, rack-scale design that unifies 72 NVIDIA Blackwell Ultra GPUs and 36 Arm®-based NVIDIA Grace CPUs in a single platform optimized for test-time scaling inference. [Learn More About GB300 NVL72](https://www.nvidia.com/en-us/data-center/gb300-nvl72.md) ### NVIDIA NVLink NVIDIA NVLink™ and NVLink Switch Chip enable 130 TB/s of bandwidth in one 72-accelerator NVLink domain (NVL72). They also deliver 4x bandwidth efficiency with NVIDIA Scalable Hierarchical Aggregation and Reduction Protocol (SHARP)™ FP8 support. [Learn More About NVLink](https://www.nvidia.com/en-us/data-center/nvlink.md) ## Programs and Technical Training ### NVIDIA Program for Startups NVIDIA Inception provides over 15,000 worldwide members with access to the latest developer resources, preferred pricing on NVIDIA software and hardware, and exposure to the venture capital community. The program is free and available to tech startups of all stages. [Learn More and Apply](https://www.nvidia.com/en-us/startups.md) ### DGX Administrator Training Learn how to administer the NVIDIA DGX™ platform for all clusters and systems. Unique courses for [DGX H100](https://www.nvidia.com/en-us/data-center/dgx-h100.md) and [A100](https://www.nvidia.com/en-us/data-center/dgx-a100.md), [DGX BasePOD](https://www.nvidia.com/en-us/data-center/dgx-basepod.md), [DGX SuperPOD](https://www.nvidia.com/en-us/data-center/dgx-superpod.md), and even [DGX Cloud](https://www.nvidia.com/en-us/data-center/dgx-cloud.md) offer attendees the knowledge to administer and deploy the platform successfully. [View Courses](https://academy.nvidia.com/en/training-by-topic/?training_by_topic=64) ### NVIDIA Training Our expert-led courses and workshops provide learners with the knowledge and hands-on experience necessary to unlock the full potential of NVIDIA solutions. Our customized training plans are designed to bridge technical skill gaps and provide relevant, timely, and cost-effective solutions for an organization's growth and development. [Visit NVIDIA Training](https://www.nvidia.com/en-us/learn/organizations.md) ## Meet Our Partners Register now to join NVIDIA at OCP. [Register Now](https://www.opencompute.org/summit/global-summit/registration)