kubeflow/trainer

Enhanced MPI Support in TrainJob

Chiusa

#2751 aperta il 24 lug 2025

 (26 commenti) (3 reazioni) (0 assegnatari)Go (964 fork)github user discovery
area/controllerarea/mpiarea/runtimesgood first issuehelp wantedkind/feature

Metriche repository

Star
 (2112 stelle)
Metriche merge PR
 (Metriche PR in attesa)

Descrizione

What you would like to be added?

We plan to enhance MPI support in TrainJob in these areas:

  • SSH-based node communication
  • Add resource per node support
  • Address the failing launcher issue in the initialization phase (launcher fails until the MPI worker is ready)

Why is this needed?

As we discussed in the WG call yesterday, this will be included in Kubeflow Trainer ROADMAP 2025.

/cc @kubeflow/wg-training-leads @astefanutti @franciscojavierarceo @kramaranya @rudeigerc @mahdikhashan @Doris-xm @Harshal292004 @saileshd1402 @kannon92 @deepanker13

Love this feature?

Give it a 👍 We prioritize the features with most 👍

Guida contributor