google / xpk

xpk (Accelerated Processing Kit, pronounced x-p-k,) is a software tool to help Cloud developers to orchestrate training jobs on accelerators such as TPUs and GPUs on GKE.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

List of Nits From An Early User

rwitten opened this issue · comments

  • XPK didn't work on Debian because awk -e wasn't supported.
  • 40 char limit for workloads is constraining, any ideas to fix.
  • Deleting workloads periodically automatically?
  • Pausing jobs
  • Upload HLO in background. (Won't fix for now, the HLO uploader we have is a short term hack before you can directly upload it to GCS)