Open Compute Project Courses

SONiC Orchagent Event-Based Retry Logic Redesign

Discover how Alibaba engineers redesigned SONiC's orchagent retry logic to prevent workflow blocking and improve network system performance through intelligent event-based strategies.

• YouTube
• 22 minutes
• Self-Paced
• Free Video

Leveraging Digital Infrastructure in Manufacturing for Performance Optimization

Discover how advanced digital infrastructure transforms manufacturing data into actionable insights for energy reduction and sustainability optimization.

• YouTube
• 20 minutes
• Self-Paced
• Free Video

The Immediate Impact of Open Networking - The Rakuten Use Case

Discover how Rakuten leveraged open networking with OCP whiteboxes and SONiC to reduce vendor dependency, improve business resiliency, and cut costs while scaling operations.

• YouTube
• 23 minutes
• Self-Paced
• Free Video

OCP Aligned Seawater Cooling and Heat Reuse Enabling Circular Infrastructure for High Density

Explore innovative seawater cooling systems for high-density data centers that transform waste heat into potable water through membrane distillation technology.

• YouTube
• 21 minutes
• Self-Paced
• Free Video

SONiC Flaky Test Case Management System

Discover how to manage flaky test cases in SONiC's 1,000+ nightly tests using automated failure detection, categorization, and prioritization systems for better debugging workflows.

• YouTube
• 35 minutes
• Self-Paced
• Free Video

Progress in Optical Interconnects for AI Clusters

Discover how optical interconnects are revolutionizing AI cluster performance through expert insights from industry leaders at NVIDIA, Arista, TE Connectivity, and Linktel Technologies.

• YouTube
• 31 minutes
• Self-Paced
• Free Video

From Printing Press to Superintelligence - Infrastructure for the Next Great Unknown

Explore hyperscale liquid cooling deployment strategies, wet vs dry shipping decisions, and industry standardization challenges with executives from Meta, NVIDIA, Google, and Microsoft.

• YouTube
• 36 minutes
• Self-Paced
• Free Video

Cooling High Power Racks (1 MW+)

Discover thermal management strategies for cooling high-power data center racks exceeding 1 MW with industry experts from Dell, AMD, Microsoft, and Vertiv.

• YouTube
• 30 minutes
• Self-Paced
• Free Video

Applications and Use Cases for UALink - Enabling Efficient Open Networking for AI Scale-Up Systems

Explore UALink's real-world applications for AI scale-up systems, covering high-performance interconnects, supply chain diversity, and next-generation AI-ML infrastructure development.

• YouTube
• 33 minutes
• Self-Paced
• Free Video

Drive Regeneration in Action - Enhancing Fault Tolerance in Datacenters

Explore innovative HDD technology that keeps drives with defects operational, expanding datacenter fault tolerance and reducing service personnel deployments.

• YouTube
• 23 minutes
• Self-Paced
• Free Video

Manage and Secure Your PCIe Infrastructure Using eSFI and PCIe MI

Discover how to leverage eSFI and PCIe MI standards for hardware management, enabling BMCs to control PCIe components independently of host operating systems.

• YouTube
• 21 minutes
• Self-Paced
• Free Video

INC Manager for AI Fabrics - Orchestrating In-Network Collective Acceleration in AI Infrastructure

Explore In-Network Collective (INC) technology and management systems for AI fabric infrastructure, covering switch-based acceleration and SONiC deployment architectures.

• YouTube
• 22 minutes
• Self-Paced
• Free Video

AVS for Vendor Simulation and Topologies with Broadcom and Arista

Explore Alpine Virtual Switch integration with Broadcom BCMSIM and Arista for SONiC testing, enabling virtual ASIC simulation and large-scale network deployment validation.

• YouTube
• 29 minutes
• Self-Paced
• Free Video

Precision Time Protocol for AI Back End Networks - Modeling and Deployment Insights

Explore Precision Time Protocol implementation for AI networks, focusing on time synchronization optimization for GPU clusters and large language model infrastructures.

• YouTube
• 29 minutes
• Self-Paced
• Free Video

Advancing Timing Resilience in Edge to Core Systems - From White Rabbit to PCIe

Discover nanosecond-level synchronisation solutions for AI workloads using ultra-stable oscillators, precision timing distribution, and advanced clock references for data centres.

• YouTube
• 31 minutes
• Self-Paced
• Free Video