You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Angel Conde (JIRA)" <ji...@apache.org> on 2019/01/07 13:56:00 UTC
[jira] [Commented] (SPARK-24725) Discuss necessary info and access
in barrier mode + Mesos
[ https://issues.apache.org/jira/browse/SPARK-24725?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16735836#comment-16735836 ]
Angel Conde commented on SPARK-24725:
--------------------------------------
Mesos has already the possibility via MPI framework ([https://github.com/apache/mesos/tree/master/mpi]). However that implementation is 5 year old (and seems to be a proof of concept) and I do not know whether that approach could be used using Docker images as executors.
Bests
> Discuss necessary info and access in barrier mode + Mesos
> ---------------------------------------------------------
>
> Key: SPARK-24725
> URL: https://issues.apache.org/jira/browse/SPARK-24725
> Project: Spark
> Issue Type: Story
> Components: ML, Spark Core
> Affects Versions: 3.0.0
> Reporter: Xiangrui Meng
> Priority: Major
>
> In barrier mode, to run hybrid distributed DL training jobs, we need to provide users sufficient info and access so they can set up a hybrid distributed training job, e.g., using MPI.
> This ticket limits the scope of discussion to Spark + Mesos. I'm not aware of MPI support in Mesos. So we should find someone with good knowledge to lead the discussion here.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org