Open Compute Project Courses

Driving Innovation in Network Infrastructure - Key Advances from the Sonic Routing Workgroup

Discover key advances in network infrastructure through SRv6, Phoenixwing, VxLAN Multi-Homing, FRR upgrades, and RIB-FIB separation architecture from the Sonic Routing workgroup.

• YouTube
• 21 minutes
• On-Demand
• Free Video

SAI Specifications for Newer Technologies Introduced by Ultra Ethernet Consortium

Explore SAI specifications for next-gen AI-ML fabric technologies including Layer 2 Ethernet enhancements, link layer retry, credit-based flow control, and network telemetry.

• YouTube
• 23 minutes
• On-Demand
• Free Video

Enabling Technologies for Next Generation Large Scale AI Backend Networking

Explore Microsoft's next-generation AI backend network architecture featuring SRv6, High-Frequency Streaming Telemetry, and trimming for hyper-scale GPU clusters.

• YouTube
• 26 minutes
• On-Demand
• Free Video

RSR Rapid Hitful Software Upgrade - Enhancing SONiC Network Switch Upgrades

Explore rapid hitful software upgrade techniques in SONiC to minimize network downtime while maintaining simplicity and enabling faster release cycles.

• YouTube
• 27 minutes
• On-Demand
• Free Video

Accurate Cable Length Measurement for Optimized Data Center Networking

Discover precise cable measurement techniques to optimize data center network performance and reduce infrastructure costs through improved planning.

• YouTube
• 21 minutes
• On-Demand
• Free Video

Standardizing Carbon Assessment Across the Industry

Explore how cloud industry leaders from Google, Meta, and Microsoft collaborate on standardizing carbon assessment methods to drive sustainability and net-zero goals.

• YouTube
• 35 minutes
• On-Demand
• Free Video

AI-Assisted Test Debug Flow and Log Analysis for AI GPU Systems

Discover AI-powered debugging workflows for GPU datacenters using machine learning to analyze logs, identify error patterns, and accelerate root cause analysis in modern AI systems.

• YouTube
• 21 minutes
• On-Demand
• Free Video

Establishing Common Redfish Requirements with a Baseline Profile

Discover how the OCP Baseline Redfish Service Profile establishes universal management requirements across diverse data center equipment types and hardware designs.

• YouTube
• 24 minutes
• On-Demand
• Free Video

OpenRMC - Increasing the Role of the Rack Manager in Data Center Management

Explore how rack managers evolve beyond proxies to intelligent controllers with AI integration, predictive failure detection, and automated decision-making for next-gen data centers.

• YouTube
• 25 minutes
• On-Demand
• Free Video

From AC to DC - The Future of High Density, AI Ready Data Centers

Explore how DC power architectures like +--400VDC can revolutionize AI-ready data centers, delivering 10-15% energy savings and supporting unprecedented 1MW per rack densities.

• YouTube
• 25 minutes
• On-Demand
• Free Video

Introducing SONiC Community Hardware Testing Lab

Discover how the new SONiC Community Hardware Testing Lab at UNH improves network switch software quality through automated hardware testing and community collaboration.

• YouTube
• 31 minutes
• On-Demand
• Free Video

Clean Standby Power Solutions - Powering the Net Zero Transition Beyond Backup

Explore clean backup power solutions for data centers, covering grid stabilization, alternative technologies, AI load profiles, and the transition to net zero energy systems.

• YouTube
• 37 minutes
• On-Demand
• Free Video

Immersion Cooling in AI Data Centers - Insights from a Multi-Year Deployment

Discover operational insights from multi-year immersion cooling deployment in AI data centers, covering hardware compatibility, fluid aging, system reliability, and performance optimization.

• YouTube
• 21 minutes
• On-Demand
• Free Video

INSPECT - Data Analysis Tool for Proactive Link Failure Detection on Meta's AI Infrastructure

Discover INSPECT, Meta's ML-powered tool for predicting link failures in AI infrastructure by analyzing SerDes parameters across 100K+ clusters to prevent downtime.

• YouTube
• 25 minutes
• On-Demand
• Free Video

Standardizing GPU Management - Redfish, Telemetry, and Firmware Update Protocols

Discover how major tech companies standardize GPU management through Redfish protocols, telemetry interfaces, and firmware updates for scalable AI and HPC workloads.

• YouTube
• 28 minutes
• On-Demand
• Free Video