Hypercomputer Clusters: The secret to AI training at Google scale

61
Published on 24 Mar 2025, 23:22
See the detailed reference architecture → goo.gle/4isJtkc

Learn how to use Hypercomputer Cluster to train models using physically co-located accelerators, targeted workload placement, advanced maintenance controls to minimize workload disruption, and topology-aware scheduling.

Subscribe to Google Cloud Tech → goo.gle/GoogleCloudTech

Speaker: Don McCasland
Products Mentioned: AI Infrastructure
autotechmusickids