You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "junbiao chen (Jira)" <ji...@apache.org> on 2021/03/10 03:18:00 UTC
[jira] [Commented] (FLINK-19938) Implement shuffle data read
scheduling for sort-merge blocking shuffle
[ https://issues.apache.org/jira/browse/FLINK-19938?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17298502#comment-17298502 ]
junbiao chen commented on FLINK-19938:
--------------------------------------
Hi! Shuffle IO scheduling is a good idea for improving IO peformance.But there is a point need to consider,which is Linux IO scheduler. When linux receive IO request,Linux IO scheduler will schedule these IO requests,IO requests will be merged and sorted.Linux‘s IO requests sorting seems to have done what Shuffle IO scheduling can do.
> Implement shuffle data read scheduling for sort-merge blocking shuffle
> ----------------------------------------------------------------------
>
> Key: FLINK-19938
> URL: https://issues.apache.org/jira/browse/FLINK-19938
> Project: Flink
> Issue Type: Sub-task
> Components: Runtime / Network
> Reporter: Yingjie Cao
> Priority: Major
> Labels: pull-request-available
> Fix For: 1.13.0
>
>
> As described in [https://cwiki.apache.org/confluence/display/FLINK/FLIP-148%3A+Introduce+Sort-Merge+Based+Blocking+Shuffle+to+Flink.] shuffle IO scheduling is important for performance. We'd like to Introduce it to sort-merge shuffle first.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)