flink-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "ASF GitHub Bot (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (FLINK-5134) Aggregate ResourceSpec for chained operators when generating job graph
Date Mon, 13 Mar 2017 13:11:41 GMT

    [ https://issues.apache.org/jira/browse/FLINK-5134?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15907424#comment-15907424

ASF GitHub Bot commented on FLINK-5134:

Github user zhijiangW commented on the issue:

    @StephanEwen , I have submitted the modifications as we confirmed before. It needs to
change some previous codes and excuse my late update because of a little busy last week.
    BTW, I found another potential bug in `StreamingJobGraphGeneratorTest.testChainStartEndSetting`.
If this test is not executed at first, it would cause NPE problem because the `idCounter`
in `StreamTransformation` is static. And I want to fix it in another separate jira although
I have modified it in order to confirm the `clean verify` successful.

> Aggregate ResourceSpec for chained operators when generating job graph
> ----------------------------------------------------------------------
>                 Key: FLINK-5134
>                 URL: https://issues.apache.org/jira/browse/FLINK-5134
>             Project: Flink
>          Issue Type: Sub-task
>          Components: DataStream API
>            Reporter: zhijiang
>            Assignee: zhijiang
> This is a part of fine-grained resource configuration in flip-6.
> In *JobGraph* generation, each *JobVertex* may contain a series of chained operators,
and the resource of *JobVertex* should be aggregation of individual resource in chained operators.
> For memory resource in *JobVertex*, the aggregation is the sum formula for chained operators,
and for cpu cores resource, the aggregation is the maximum formula for chained operators.

This message was sent by Atlassian JIRA

View raw message