[
https://issues.apache.org/jira/browse/MAPREDUCE-4946?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13556391#comment-13556391
]
Jason Lowe commented on MAPREDUCE-4946:
---------------------------------------
This performance problem prevents the AM from reliably supporting very large
jobs (i.e.: tens of thousands of maps and thousands of reducers) because it can
take too long to serve up requests and other clients end up being ignored and
timeout. If the same task times out enough attempts then the whole job fails.
> Type conversion of map completion events leads to performance problems with
> large jobs
> --------------------------------------------------------------------------------------
>
> Key: MAPREDUCE-4946
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-4946
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Components: mr-am
> Affects Versions: 2.0.2-alpha, 0.23.5
> Reporter: Jason Lowe
> Priority: Critical
>
> We've seen issues with large jobs (e.g.: 13,000 maps and 3,500 reduces) where
> reducers fail to connect back to the AM after being launched due to
> connection timeout. Looking at stack traces of the AM during this time we
> see a lot of IPC servers stuck waiting for a lock to get the application ID
> while type converting the map completion events. What's odd is that normally
> getting the application ID should be very cheap, but in this case we're
> type-converting thousands of map completion events for *each* reducer
> connecting. That means we end up type-converting the map completion events
> over 45 million times during the lifetime of the example job (13,000 * 3,500).
> We either need to make the type conversion much cheaper (i.e.: lockless or at
> least read-write locked) or, even better, store the completion events in a
> form that does not require type conversion when serving them up to reducers.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
