You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@ignite.apache.org by "Vladimir Ozerov (JIRA)" <ji...@apache.org> on 2016/12/19 14:59:58 UTC

[jira] [Created] (IGNITE-4461) Hadoop: automatically define best "partially raw" comparator based on key type.

Vladimir Ozerov created IGNITE-4461:
---------------------------------------

             Summary: Hadoop: automatically define best "partially raw" comparator based on key type.
                 Key: IGNITE-4461
                 URL: https://issues.apache.org/jira/browse/IGNITE-4461
             Project: Ignite
          Issue Type: Task
          Components: hadoop
    Affects Versions: 1.8
            Reporter: Vladimir Ozerov
            Assignee: Vladimir Ozerov
             Fix For: 2.0


We have special {{PartiallyRawComparator}} interface which provides nice speedup by removing unnecessary objects deserializations during insert into result multi-map. And we have a one implementation for now for {{Text}} class: {{TextPartiallyRawComparator}}.

We should do the following:
1) Try guessing proper comparator in runtime using {{JobConf.getOutputKeyClass()}} property. 
2) Add ability to disable this automatic resolution.



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