spark-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From rxin <...@git.apache.org>
Subject [GitHub] spark pull request: [SPARK-5843] Allowing map-side combine to be s...
Date Mon, 09 Mar 2015 17:37:32 GMT
Github user rxin commented on the pull request:

    https://github.com/apache/spark/pull/4634#issuecomment-77903115
  
    Serializer seems ok to add.
    
    One thing I am not sure about is the mapSideCombine thing -- I'm never a fan of that parameter
even though I added it myself, for the following reasons:
    
    1. mapSideCombine is a MR term used in Hive that doesn't mean much outside of MR. A more
proper name is partialAggregation.
    2. The underlying implementation should be able to avoid partial aggregation if it finds
that partial aggregation is expensive (i.e. after trying 10000 records, check whether the
hash table size is less than a specific threshold). It is one of the things we can easily
auto tune. 



---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org


Mime
View raw message