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.
Apache License 2.0
69
stars
17
forks
source link
Allow JAX coordinator to find the JobSet name. #140
CPU based workloads are not able to find the JAX coordinator. Reordering of env vars is needed following - https://github.com/google/xpk/pull/124 , as JOBSET_NAME comes through the env_vars and that is being used in the JAX_COORDINATOR.
Fixes / Features
Testing / Documentation
Testing details.