GitHub user navis opened a pull request:
https://github.com/apache/spark/pull/6790
[SPARK-8334] [SQL] Binary logical plan should provide more realistic
statistics
Currently, spark-sql multiplies size of two children for output size, which
makes following join hash the other side of input. In multi way joins something
like (A join B) join C, C will be marked as 'buildHash' target if size of C is
not bigger than (size of A * size of B). Some results of TPC-H queries are
greatly affected by this.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/navis/spark SPARK-8334
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/spark/pull/6790.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 #6790
----
commit 2afe381cb5e9c7bfeba6219a1cf2b3fd1eeebfe0
Author: navis.ryu <[email protected]>
Date: 2015-06-12T19:46:16Z
[SPARK-8334] [SQL] Binary logical plan should provide more realistic
statistics
----
---
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]
