hadoop-mapreduce-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Binglin Chang (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (MAPREDUCE-2841) Task level native optimization
Date Mon, 29 Aug 2011 17:04:38 GMT

    [ https://issues.apache.org/jira/browse/MAPREDUCE-2841?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13092972#comment-13092972
] 

Binglin Chang commented on MAPREDUCE-2841:
------------------------------------------

bq. It might make sense to commit that subset as optional functionality first, then iterate
based on feedback.
I agree. How to contribute this to hadoop? Add a new subdirectory in contrib like streaming,
or merge to native, or stay in current c++/libnativetask?
It contains both c++ and java code, and will likely to add client tools like streaming, and
dev SDK.

Random memory config gives the Resource Scheduler more information so it may yield better
schedule algorithms. As for OOM, there is a flex layer for memory control already, page cache.
In typical slave node memory configuration and real cases, page cache (should) take considerable
proportions of total memory(20%-50%), so for example tasks can be configured to use 60% of
memory, but can have some variance in 20% range, and the variance become relatively small
when multiple tasks combined to node level or whole job level.
One of my colleague is working on shuffle service, which delegate all reduce shuffle work
to a per node service, this has some aspect which is similar:
For a single task, the variance of memory footprint is a problem, but it gets much stable
for many tasks run on a node.


> Task level native optimization
> ------------------------------
>
>                 Key: MAPREDUCE-2841
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-2841
>             Project: Hadoop Map/Reduce
>          Issue Type: Improvement
>          Components: task
>         Environment: x86-64 Linux
>            Reporter: Binglin Chang
>            Assignee: Binglin Chang
>         Attachments: MAPREDUCE-2841.v1.patch, dualpivot-0.patch, dualpivotv20-0.patch
>
>
> I'm recently working on native optimization for MapTask based on JNI. 
> The basic idea is that, add a NativeMapOutputCollector to handle k/v pairs emitted by
mapper, therefore sort, spill, IFile serialization can all be done in native code, preliminary
test(on Xeon E5410, jdk6u24) showed promising results:
> 1. Sort is about 3x-10x as fast as java(only binary string compare is supported)
> 2. IFile serialization speed is about 3x of java, about 500MB/s, if hardware CRC32C is
used, things can get much faster(1G/s).
> 3. Merge code is not completed yet, so the test use enough io.sort.mb to prevent mid-spill
> This leads to a total speed up of 2x~3x for the whole MapTask, if IdentityMapper(mapper
does nothing) is used.
> There are limitations of course, currently only Text and BytesWritable is supported,
and I have not think through many things right now, such as how to support map side combine.
I had some discussion with somebody familiar with hive, it seems that these limitations won't
be much problem for Hive to benefit from those optimizations, at least. Advices or discussions
about improving compatibility are most welcome:) 
> Currently NativeMapOutputCollector has a static method called canEnable(), which checks
if key/value type, comparator type, combiner are all compatible, then MapTask can choose to
enable NativeMapOutputCollector.
> This is only a preliminary test, more work need to be done. I expect better final results,
and I believe similar optimization can be adopt to reduce task and shuffle too. 

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message