Skip to content

Do we have to do tracing in a real system when we try to analyze a 10,000 GPUs training system? #739

Do we have to do tracing in a real system when we try to analyze a 10,000 GPUs training system?

Do we have to do tracing in a real system when we try to analyze a 10,000 GPUs training system? #739

Triggered via issue November 25, 2024 02:06
@191220042191220042
commented on #164 b915ab8
Status Success
Total duration 11s
Artifacts

cla.yml

on: issue_comment
cla-check
1s
cla-check
Fit to window
Zoom out
Zoom in