264001
eyJ0eXAiOiJKV1QiLCJhbGciOiJIUzI1NiJ9.eyJpc3MiOiJid2V2ZW50cy1wdWJsaWMtYXBpIiwiYXVkIjpbInB1YmxpYy13ZWIiXSwic3ViIjoiaHR0cHM6Ly93d3cuZ29vZ2xlY2xvdWRldmVudHMuY29tIiwiaWF0IjoxNzc3MzgzOTA2LCJuYmYiOjE3NzczODM5MDYsImV4cCI6MTc3NzM4NzUwNiwiY29udGV4dCI6eyJwYXlsb2FkIjp7ImlkIjoiIiwicmVnVHlwZU5hbWUiOiIiLCJldmVudF9pZCI6IjI2NDAwMSIsImNfMzI5MDQxNiI6IiJ9fX0.XOGXBUTrNgdv1Y1popcd4XA1CIoSw6YSYc7MNrYoLpU
phase 7 playlists session info modal
Day 3 – April 24, 2026
Discussion Groups
BRK3-029A-DG • Kubernetes, Cloud Runtimes • Advanced Technical
Building an AI supercomputer: Scale Slurm workloads on GKE
location_on Reef A
schedule 12:30 PM - 1:00 PM

Dive into the architecture of running Slurm on GKE for large-scale AI training. We will detail the new Managed Slinky architecture and its deep integration with Google’s AI Hypercomputer stack—including Cluster Director and native TPU support. Learn how to combine the raw throughput of bare-metal HPC with the operational simplicity of GKE to accelerate your model training velocity.

 Read more
This agenda widget contains logic for handling same sessions available at multiple times
GTM Tracking Header name present
Session
contains logic for gtm tracking when modal open