kubeflow/trainer

Enhanced MPI Support in TrainJob

已关闭

#2,751 创建于 2025年7月24日

 (26 条评论) (3 个反应) (0 位负责人)Go (964 个派生)github user discovery
area/controllerarea/mpiarea/runtimesgood first issuehelp wantedkind/feature

仓库指标

星标
 (2,112 个星标)
PR 合并指标
 (PR 指标待抓取)

描述

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 👍

贡献者指南