GitHub user JoshRosen opened a pull request:
https://github.com/apache/spark/pull/7334
[SPARK-8964] Use Exchange to perform shuffle in Limit
Spark SQL's physical Limit operator currently performs its own shuffle
rather than using Exchange to perform the shuffling. This is less efficient
since this non-exchange shuffle path won't be able to benefit from SQL-specific
shuffling optimizations, such as SQLSerializer2. It also involves additional
unnecessary row copying.
Instead, I think that we should rewrite Limit to expand into three physical
operators:
PerParititonLimit -> Exchange to one partition -> PerPartitionLimit
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/JoshRosen/spark remove-copy-in-limit
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/spark/pull/7334.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #7334
----
commit dfe6ff100ae2efcb1778d93bf828f9e2e8c46e18
Author: Josh Rosen <[email protected]>
Date: 2015-07-09T22:27:05Z
Use Exchange in Limit (SPARK-8964)
----
---
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 [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
