Completed
0:00 Benchmarking Google’s TPUs vs Nvidia GPUs
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Benchmarking Google's TPUs vs Nvidia GPUs for AI Inference
Automatically move to the next video in the Classroom when playback concludes
- 1 0:00 Benchmarking Google’s TPUs vs Nvidia GPUs
- 2 0:33 Video Overview
- 3 1:12 H100 SXM, H200 SXM and v6e hardware specs
- 4 4:47 Benchmarking Design with vLLM and llmperf
- 5 7:42 Price assumptions per hour
- 6 8:47 Tensor Parallel vs Pipeline Parallel
- 7 13:45 Pros and Cons of Tensor vs Pipeline Parallel
- 8 14:42 Where to test TPUs and GPUs
- 9 15:45 Future videos: Blackwell B200 and Amazon Trainium
- 10 16:15 Running inference on Nvidia GPUs
- 11 19:17 Running inference on Google TPUs
- 12 25:51 Running benchmarking with llmperf
- 13 28:23 Benchmarking Results: TPU vs GPU
- 14 33:21 Conclusion, Resources and Workshop
