Unlocking ML Performance on TPUs and GPUs

Maximize the efficiency of your ML workloads on Google Cloud. This session delivers strategies for in-depth diagnostics on TPUs and GPUs. Learn to analyze critical metrics and achieve performance through practical diagnostic techniques.

Speakers: Rajesh Anantharaman, Niranjan Hira

Check out the cloud session track from Google I/O 2025 → https://goo.gle/io25-gct-yt
Check out all of the sessions from Google I/O 2025→ https://goo.gle/io25-sessions-yt

Subscribe to Google Cloud Tech → https://goo.gle/GoogleCloudTech

Event: Google I/O 2025

Products Mentioned: AI/Machine Learning

Видео Unlocking ML Performance on TPUs and GPUs канала Google Cloud Tech