[
https://issues.apache.org/jira/browse/PIG-2228?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Thejas M Nair updated PIG-2228:
-------------------------------
Attachment: PIG-2228.4.patch
This patch adds auto-disabling of in-map partial aggregation if there is not
sufficient reduction in map records. The threshold for minimum reduction that
should happen is set using the property pig.exec.mapPartAgg.minReduction .
With minimum auto disabling threshold set to 10 (ie 10 input map records to one
output record), the queries which had performance degradation with this feature
turned on now performed at par.
For now, the feature is turned off by default, we can look into changing this
after users get more time to try it out.
> support partial aggregation in map task
> ---------------------------------------
>
> Key: PIG-2228
> URL: https://issues.apache.org/jira/browse/PIG-2228
> Project: Pig
> Issue Type: Bug
> Reporter: Thejas M Nair
> Assignee: Thejas M Nair
> Fix For: 0.10
>
> Attachments: PIG-2228.1.patch, PIG-2228.2.patch, PIG-2228.3.patch,
> PIG-2228.4.patch
>
>
> h3. Introduction
> Pig does (sort based) partial aggregation in map side through the use of
> combiner. MR serializes the output of map to a buffer, sorts it on the keys,
> deserializes and passes the values grouped on the keys to combiner phase. The
> same work of combiner can be done in the map phase itself by using a hash-map
> on the keys. This hash based (partial) aggregation can be done with or
> without a combiner phase.
> h3. Benefits
> It will send fewer records to combiner and thereby -
> * Save on cost of serializing and de-serializing
> * Save on cost of lock calls on the combiner input buffer. (I have found
> this to be a significant cost for a query that was doing multiple group-by's
> in a single MR job. -Thejas)
> * The problem of running out of memory in reduce side, for queries like
> COUNT(distinct col) can be avoided. The OOM issue happens because very large
> records get created after the combiner run on merged reduce input. In case of
> combiner, you have no way of telling MR not to combine records in reduce
> side. The workaround is to disable combiner completely, and the opportunity
> to reduce map output size is lost.
> * When the foreach after group-by has both algebraic and non-algebraic
> functions, or if a bag is being projected, the combiner is not used. This is
> because the data size reduction in typical cases are not significant enough
> to justify the additional (de)serialization costs. But hash based aggregation
> can be used in such cases as well.
> * It is possible to turn off the in-map combine automatically if there is
> not enough 'combination' that is taking place to justify the overhead of the
> in-map combiner. (Idea borrowed from Hive jira.)
> * If input data is sorted, it is possible to do efficient map side
> (partial) aggregation with in-map combiner.
> Design proposal is here -
> https://cwiki.apache.org/confluence/display/PIG/PigInMapCombinerProposal
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
