You are viewing a plain text version of this content. The canonical link for it is here.
Posted to mapreduce-issues@hadoop.apache.org by "jaehoon ko (JIRA)" <ji...@apache.org> on 2014/06/27 03:16:25 UTC
[jira] [Created] (MAPREDUCE-5947) Map phase merge can better
utilize memory
jaehoon ko created MAPREDUCE-5947:
-------------------------------------
Summary: Map phase merge can better utilize memory
Key: MAPREDUCE-5947
URL: https://issues.apache.org/jira/browse/MAPREDUCE-5947
Project: Hadoop Map/Reduce
Issue Type: Improvement
Components: performance
Affects Versions: 2.4.0
Reporter: jaehoon ko
Assignee: jaehoon ko
Map phase merge reads spills from disk and writes intermediate results back to disk, and so on. I think it is possible to use memory to store intermediate results, thereby reducing disk IO. Because kvbuffer is nullified right before merge, we have at least io.sort.mb amount of heap available.
MAPREDUCE-4511 can be considered as an effort to utilize memory better through read ahead, but number of disk IO is unchanged.
Please give me your thoughts. I'd like to take up this issue.
--
This message was sent by Atlassian JIRA
(v6.2#6252)