Explore a 21-minute conference presentation from industry experts at Broadcom, NVIDIA, and Dell Technologies discussing the latest advancements in the SONiC AI Working Group. Gain deep insights into crucial developments in network fabrics for AI, routing and network design, congestion control, and load balancing. Discover how SONiC is being enhanced to support AI clusters, understand current technological gaps, and learn about upcoming developments in the roadmap. Examine the evolution of SONiC as it adapts to meet the requirements of high-performance, low-latency network architectures specifically designed for AI clusters and GPU-accelerated workloads.
Advancing SONiC for AI - Network Architecture and Performance Optimization
Open Compute Project via YouTube
Overview
Syllabus
Advancing SONiC for AI Insights from the Latest Developments in the SONiC AI Working Group
Taught by
Open Compute Project
Related Courses
-
Empowering AI Networking with SONiC - Building Robust Network Infrastructure
-
Insights from Production Scheduled Ethernet Fabric in Large AI Training Clusters
-
Building the First SONiC Cloud AI Benchmarked Cluster
-
Scheduled Ethernet Fabric for Large-Scale AI Training Clusters
-
Next Generation Fabric for AI/ML Use Cases - Disaggregated Scheduled Fabric Using SONiC
-
Google Cloud Network Infrastructure for AI and Machine Learning
