GitHub user viirya opened a pull request:
https://github.com/apache/spark/pull/8226
[SPARK-10031][SQL] Join two UnsafeRows in SortMergeJoin if possible
JIRA: https://issues.apache.org/jira/browse/SPARK-10031
Currently in `SortMergeJoin`, when two rows from left and right plans are
both `UnsafeRow`, we still use `JoinedRow` to join them and do an extra
`UnsafeProjection` later.
We can just use `GenerateUnsafeRowJoiner` to join two `UnsafeRows` in
`SortMergeJoin` if possible. Besides, `GenerateUnsafeRowJoiner` can have a
`withRight` function to only update row2 with a same row1.
Add `withRight` function to `GenerateUnsafeRowJoiner`. We can avoid another
unnecessary data copying for row1.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/viirya/spark-1 join-unsaferow-smj
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/spark/pull/8226.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 #8226
----
commit f4ae0cef4c5a67700748176156eafa3478d45b55
Author: Liang-Chi Hsieh <[email protected]>
Date: 2015-08-15T18:39:45Z
Join two UnsafeRows in SortMergeJoin if possible. Add withRight function to
GenerateUnsafeRowJoiner.
----
---
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]
