Open Compute Project Courses

Device Security in the OCP Supply Chain - Overview of How OCP SAFE Addresses Device Firmware Challenges

Explore how OCP S.A.F.E. framework tackles device firmware security challenges and strengthens supply chain integrity through expert insights and practical solutions.

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

PCIe Active Electrical Cables (AECs) for Large Language Model (LLM) Computing Clusters

Explore the evolution and implementation of PCIe Active Electrical Cables (AECs) for GPU clustering in Large Language Models, focusing on retimer technology advantages and form factor considerations.

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

Implementing Out-of-Band CPER Logs in Meta's AI ML System - A Case Study

Dive into Meta's implementation of standardized CPER error reporting for AI accelerators, exploring challenges and solutions for robust out-of-band hardware monitoring systems.

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

Debugging Capabilities and Best Practices for AI GPU Systems

Explore debugging challenges and best practices for AI GPU systems, focusing on PCIe subsystems, system crashes, and management controllers while learning essential troubleshooting tools and techniques.

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

Evolving FBOSS for the Next Generation AI Network Fabric

Discover how Meta evolved FBOSS to handle unique AI Fabric networking challenges, focusing on elephant flows management, hash collision mitigation, and dataplane/control plane adaptations.

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

Congestion Management in Ethernet-Based Networks for AI Cluster Fabric

Explore advanced congestion control methods for AI cluster fabrics, focusing on endpoint and fabric scheduling to optimize network performance, minimize latency, and maximize throughput.

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

DPU-Accelerated Petascale Storage Subsystem for Near Data Computing

Explore how Samsung's Petabyte SSD and MangoBoost's DPU technology revolutionize data-centric AI infrastructure, optimizing performance and efficiency for large-scale AI workloads.

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

Effective Data Tiering and Memory Allocation Techniques in Heterogeneous Memory System

Discover advanced techniques for optimizing data tiering and memory allocation in heterogeneous systems, focusing on efficient resource management and performance enhancement.

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

Vector Storage for Large Scale Multimodal RAG

Explore advanced vector storage techniques for implementing large-scale multimodal retrieval-augmented generation systems and their practical applications.

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

Overcoming the GPU Memory Capacity Limitation with Two-Digit Nanosecond Latency CXL Controller

Explore advanced solutions for GPU memory limitations through CXL controller technology, focusing on nanosecond-level latency optimization and enhanced capacity management.

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

Infrastructure Software for Hyperscale AI and GPU Clusters

Explore cutting-edge infrastructure software solutions powering large-scale AI operations and GPU cluster management for hyperscale computing environments.

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

OCP Hyperscale-Compliant AI Storage Servers with Compression and Deduplication Accelerators

Explore cutting-edge hyperscale AI storage solutions featuring advanced compression and deduplication technologies for optimized data management in enterprise environments.

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

OCP Ready - Open Compute Project Data Center Certification

Discover the key principles and implementation strategies of OCP Ready certification through ST Digital's comprehensive overview of data center infrastructure standards.

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

Advancing CXL Memory Architecture for Enhanced Performance and Flexible Integration

Explore advanced CXL memory architecture concepts, focusing on performance optimization, integration strategies, and the latest developments in memory system design for modern computing platforms.

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

Implementing PCIe Connectivity with Optics - Optical Solutions for High-Speed Computing

Explore the latest developments in optical PCIe connectivity, focusing on protocol adaptations, new form factors, and enhanced link management for AI and high-performance computing applications.

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