Overview
Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Watch a 56-minute technical presentation exploring how Ethernet networks are evolving to meet the demands of artificial intelligence workloads. Gain insights into network topologies commonly used with AI systems, understand traffic patterns across different AI workloads, and examine the networking challenges they present. Learn about the Ultra Ethernet Consortium's role in addressing congestion issues during model training, with particular emphasis on large language model requirements. Through detailed use cases, discover how Ethernet solutions are being implemented to handle bandwidth and congestion challenges in AI environments. Join industry experts Raguraman Sundaram from Celestica and Erik Smith from Dell Technologies as they break down complex networking concepts and demonstrate how Ethernet infrastructure is adapting to support next-generation AI applications.
Syllabus
Ethernet in the Age of AI: Adapting to New Networking Challenges
Taught by
SNIAVideo