February 2-4, 2027 · San Jose
Sessions at Digital Infrastructure World 2027
36 sessions with a published abstract, grouped by track. Each page has the full abstract, the time and the room. Sessions without an abstract yet are on the full agenda.
Rack track
Designing the 1MW Rack. Programmed by OCP Tech Day.
- Open Ecosystem Can Unlock Velocity for AI DCs
- Orion: One Open Rack for More Than One Accelerator
- Applying OCP ORv3 and ORW Design Principles to Rack-Scale Arm AI Infrastructure with Supermicro
- The Open Chiplet Economy: From Chiplets to AI Systems
- Scaling Up AI Compute, Memory and Connectivity
- Building Interoperable AI Systems with Chiplets: Arm Neoverse and Cadence in an Open Ecosystem
- Securing the Open Chiplet Ecosystem: Defining a Common Security Architecture for FCSA
- Open Adapter Card (OAC): Sub-Rackscale Acceleration Lives On
- OCP NIC 2026: Gen 7 and Liquid Cooling, More Speed and Power Ahead
- From Megawatts to Tokens: Who Operates the AI Factory
- ESUN: Open Ethernet for the AI Scale-Up Era
- MRC: A Purpose-Built RDMA Transport for AI Training Networks
- SUE-T: An Ethernet Transport for Scale-Up Networking
- OCP Air vs. Liquid - Where Does Air Cooling Still Win for General Compute
- Beyond the Cold Plate: Warm Water, Retrofits, and the Open Road for Liquid Cooling
- Consolidating Liquid Cooling and Power Management onto a Single Rack Management Controller
- Scaling Hyperscale Power Delivery: Managing Aggregated and Cascaded Power Shelf Architectures for High-Density GPU Systems
- Powering the AI Rack of the Future: A Scalable 800 V DC Sidecar Architecture
- ORv3 Power Architectures for Higher-Power Workloads
- 800V Direct-to-Server and Power Shelf Solutions
- Enabling 1MW+ IT Rack Power with an Open Rack Busbar Interface
Cluster track
Architecting 1M GPU Clusters. Programmed by CloserStill with OCP.
- Current Activities in the Open Cluster Designs for AI Strategic Initiative
- Your AI Cluster's Real Bottleneck Is How Fast It Can Change
- From Hardware to Useful Capacity: Provisioning AI Infrastructure at Scale
- Beyond Physical Digital Twins: Why Logical Digital Twins Are Essential for AI Factory Design, Validation and Operations
- Node-to-Rack Efficiency for Real-World AI Inference: Arm and FuriosaAI
- Overcoming AI's Power and Space Constraints with Scale-Across Deployments
- One Rack, More Than One Accelerator: Lessons from Building an Open AI Rack
Data Center track
Building 1GW AI Data Centers. Programmed by CloserStill with OCP.
- From Grid to Chip: System Level Simulation of MVAC and LVDC Infrastructure
- Building AI Data Centers at Speed: Managing Power, Cooling and Construction Risk
- Building the DC Power Architecture for AI Data Centers: From System Requirements to Validation and Certification
- One AI Architecture Won't Fit Every Workload: Six Deployment Topologies
Energy track
Developing 10GW Pipelines. Programmed by CloserStill with OCP.
- Co-Designing the Power Envelope: Breaking Infrastructure Silos for Gigawatt-Scale AI Clusters
- When Communities Push Back: Avoiding Escalation in AI Infrastructure
- Putting AI to Work: Shifting from AI Training to AI Inference Data Centers
- Rethinking AI Data Center Power Provisioning with Distributed Energy Storage