You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@flink.apache.org by "Greg Hogan (JIRA)" <ji...@apache.org> on 2016/09/02 15:46:20 UTC

[jira] [Created] (FLINK-4571) Configurable little parallelism in Gelly drivers

Greg Hogan created FLINK-4571:
---------------------------------

             Summary: Configurable little parallelism in Gelly drivers
                 Key: FLINK-4571
                 URL: https://issues.apache.org/jira/browse/FLINK-4571
             Project: Flink
          Issue Type: Improvement
          Components: Gelly
    Affects Versions: 1.2.0
            Reporter: Greg Hogan
            Assignee: Greg Hogan
            Priority: Minor


Several Gelly library implementations support a configurable "little parallelism" which is important when scaling to large data sets. These algorithms include operators at the beginning and end which process data on the order of the original DataSet, as well as middle operators that exchange 100s or 1000s more data. The "little parallelism" should be configurable in the appropriate Gelly drivers in the flink-gelly-examples module.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)