flink-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "zhu.qing (JIRA)" <j...@apache.org>
Subject [jira] [Created] (FLINK-8526) When use some parallelism, the program will stuck in some setting.
Date Tue, 30 Jan 2018 11:37:00 GMT
zhu.qing created FLINK-8526:
-------------------------------

             Summary: When use some parallelism, the program will stuck in some setting. 
                 Key: FLINK-8526
                 URL: https://issues.apache.org/jira/browse/FLINK-8526
             Project: Flink
          Issue Type: Bug
          Components: Cluster Management, Java API, Local Runtime
    Affects Versions: 1.4.0
         Environment: 8 machines(96GB and 24 cores)  and 20 taskslot per taskmanager. twitter-2010
dataset. And parallelism setting to 80. I run my code in standalone mode. 
            Reporter: zhu.qing
         Attachments: T2AdjActiveV.java, T2AdjMessage.java

The next program will stuck at some special parallelism in some situation. In the code attached.
And I want to know the relation between parallelism and running task.When parallelism is 80.
The program will always stuck. And when parallelism is 100. everything goes well. And I found
when the parallelism equals to number of taskslots. The program is not fasted. And probably
caused network buffer not enough. How networker buffer related to parallelism



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Mime
View raw message