Join NVIDIA at the OCP Global Summit 2026
San Jose McEnery Convention Center
October 12–15
Check out this year’s Open Compute Global Summit (OCP) to see all the latest advancements and innovations in NVIDIA’s 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.
Ian Buck, VP of Hyperscale and High-Performance Computing, NVIDIA
OCP 2026 Keynote: AI Factories at Scale: Advancing Open Infrastructure for Agentic AI
Join us on Monday, October 12, at 5:45 p.m., to discover how NVIDIA and its partners are advancing the open ecosystem through OCP and shaping the path forward for AI factories at scale.
News at OCP
AI Factory MGX Ecosystem
Walk alongside an AI factory! The NVIDIA AI Factory MGX™ Ecosystem Showcase translates rack-scale engineering into an immersive, room-scale experience that reveals how power delivery, cooling, and modular infrastructure enable the creation of intelligence at scale. With more than 200 ecosystem partners already adopting MGX designs, enterprises gain a future-proof path to AI factory success.
Featured MGX Partners
Schedule at a Glance
OCP 2026 will spotlight 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
-
Tuesday
-
Wednesday
-
Thursday
Monday, October 12 | 5:45–6:15 p.m.
Ian Buck Keynote: Advancing Open Infrastructure for Agentic AI
Street Level, South Hall
Tuesday, October 13 | 2:05–2:25 p.m.
Beyond Physical Digital Twins: Why Logical Digital Twins Are Essential for AI Factory Design, Validation, and Operations
Concourse Level, 220B
Tuesday, October 13 | 2:25–2:45 p.m.
Cooling Distribution Unit Sub-Project Overview and Updates
Concourse Level, 220C
Tuesday, October 13 | 2:30–2:45 p.m.
AI Tokenomics Explained
Concourse Level, 220B
Tuesday, October 13 | 2:55–3:15 p.m.
Every Megawatt Counts: The Path to Building AI Factories With DSX MaxLPS
Concourse Level, 210AE
Tuesday, October 13 | 4:05–4:20 p.m.
NVIDIA SDK + FBOSS: Building Production-Ready AI Network Fabrics
Concourse Level, 210AE
Tuesday, October 13 | 4:05–4:20 p.m.
800 VDC AI Factories: From Architecture Alignment to Industry Deployment
Concourse Level, 210CG
Tuesday, October 13 | 4:45–5 p.m.
Flexibility the Grid Can Count On: Turning Curtailable Load Into Interconnection Capacity
Concourse Level, 210CG
Wednesday, October 14 | 8:15–9:10 a.m.
OCP Hardware Fault Management Workstreams Status and Relationship
Concourse Level, 211
Wednesday, October 14 | 8:50–9:05 a.m.
Defining End-to-End 800VDC System Performance Requirements
Concourse Level, 210CG
Wednesday, October 14 | 9:10–9:30 a.m.
LVDC: Building an Open and Evolving Ecosystem
Concourse Level, 210CG
Wednesday, October 14 | 9:15–9:30 a.m.
Revolutionizing Networking for Gigawatt AI Factories
Concourse Level, 210AE
Wednesday, October 14 | 10:05–10:20 a.m.
Fluid Flush/Fill Cart for Pre-Deployment Conditioning of Liquid-Cooled IT Systems
Concourse Level, 220C
Wednesday, October 14 | 10:40–11 a.m.
Democratizing DPU High Availability With a SONiC-Driven DASH HA Control Plane
Concourse Level, 230A
Wednesday, October 14 | 12:30–12:50 p.m.
IO Prioritization for Optimized Command Handling
Concourse Level, 212
Wednesday, October 14 | 1:35–1:55 p.m.
Designing Safe 800 VDC Systems: Protection, Fault Management, and Operations
Concourse Level, 210CG
Wednesday, October 14 | 3:15–3:35 p.m.
Integrating Solid-State Transformer to Data Center LVDC Ecosystem
Concourse Level, 210CG
Wednesday, October 14 | 3:40–4 p.m.
Grid Interconnection Requirements for Large Loads and Path to Next-Generation Data Center Power Architectures
Concourse Level, 210CG
Wednesday, October 14 | 3:40–4:05 p.m.
OBMF Specifications and Workstream Status Update
Concourse Level, 210CG
Wednesday, October 14 | 4:35–5 p.m.
From Scale to Security: Evolving GPU Management Standards in 2026
Concourse Level, 211
Thursday, October 15 | 8–8:15 a.m.
An Introduction to Attesting Hitlessly Updatable Systems
Concourse Level, 212
Thursday, October 15 | 8–8:15 a.m.
CTAM: Three Years of Bridging Hyperscalers and Suppliers Through GPU Accelerator Compliance
Lower Level, 20BCD
Thursday, October 15 | 9:05–9:30 a.m.
Considerations for Shipping and Storage of 1-Phase Cold Plate Liquid-Cooled Racks and Servers
Concourse Level, 220C
Thursday, October 15 | 9:45–10:10 a.m.
USB to Rule Them All: One Manageability Interface for AI Platforms
Concourse Level, 220B
Thursday, October 15 | 11:15–11:30 a.m.
Evaluating Heat Rejection System Configurations for AI Data Centers: A Multi-Metric Analysis
Concourse Level, 210CG
Thursday, October 15 | 12:30–12:55 p.m.
Next-Generation ML Infrastructure: The Open DC Architecture
Concourse Level, 210CG
Thursday, October 15 | 1:15–1:30 p.m.
Secure Key Distribution Protocol for SoC Hardware Blocks
Concourse Level, 212
Thursday, October 15 | 1:35–1:50 p.m.
MGX Power Rack Reference Design
Concourse Level, 230A
Thursday, October 15 | 3:10–3:30 p.m.
Commissioning Liquid-Cooled Data Centers With Computing Equipment for Operational Readiness
Concourse Level, 210CG
Thursday, October 15 | 3:10–3:30 p.m.
Deep Dive: Biomanagement for Liquid Cooling
Concourse Level, 220C
Thursday, October 15 | 3:35–4 p.m.
The AI Factory Digital Twin: Optimizing Tokens per Watt Across the Full Stack
Concourse Level, 210CG
Architectures for the Modern Data Center and AI Factory
Explore NVIDIA Solutions
Programs and Technical Training
Frequently Asked Questions About OCP Global Summit 2026
The OCP Global Summit is a premier industry event showcasing the latest advancements and innovations in open data center design, including NVIDIA’s accelerated computing platform. It features keynotes and sessions on AI, infrastructure, networking, energy efficiency, and security.
Reference: Learn more about the OCP Global Summit
The event is scheduled for October 12–15, 2026, and will be held at the San Jose McEnery Convention Center in San Jose, California.
Reference: Register for the OCP Global Summit
Ian Buck, NVIDIA’s VP of Hyperscale and High-Performance Computing, will deliver the keynote presentation on Monday, October 12, focusing on the future of open infrastructure for AI.
NVIDIA speakers will present on a wide range of topics, including AI infrastructure, advanced networking, energy efficiency, 800 VDC, liquid cooling solutions, and security for the next generation of AI factories.
You can register for the summit through the official event registration portal.
Reference: OCP Global Summit registration portal
NVIDIA Vera Rubin is NVIDIA’s pod-scale foundation for AI factories built for agentic AI to deliver the best performance per watt and lowest token cost through extreme co-design. Vera Rubin combines five purpose-built racks into a fully integrated platform operating as one massive AI supercomputer.
Reference: Explore NVIDIA Vera Rubin platform
NVIDIA MGX™ is an open modular reference architecture that enables OEMs, ODMs, and ecosystem partners to build accelerated computing systems faster. From single-node servers to rack-scale AI factories, NVIDIA MGX reduces engineering costs, shortens time to market, and delivers multigenerational compatibility for the ecosystem.
Reference: Learn more about NVIDIA MGX
Meet Our Partners
Register now to join NVIDIA at OCP.