diff options
| author | Cheng Lian <lian@databricks.com> | 2014-10-27 10:06:05 -0700 |
|---|---|---|
| committer | Michael Armbrust <michael@databricks.com> | 2014-10-27 10:06:09 -0700 |
| commit | 1d7bcc88401d66c8d17a075355acfc25a8b7615c (patch) | |
| tree | 9777ae9fa35462e45689fe5fb4751e2a495b7424 /LICENSE | |
| parent | dea302ddbd26b1f20fb8a3979bd1d8e1717479f8 (diff) | |
| download | spark-1d7bcc88401d66c8d17a075355acfc25a8b7615c.tar.gz spark-1d7bcc88401d66c8d17a075355acfc25a8b7615c.tar.bz2 spark-1d7bcc88401d66c8d17a075355acfc25a8b7615c.zip | |
[SQL] Fixes caching related JoinSuite failure
PR #2860 refines in-memory table statistics and enables broader broadcasted hash join optimization for in-memory tables. This makes `JoinSuite` fail when some test suite caches test table `testData` and gets executed before `JoinSuite`. Because expected `ShuffledHashJoin`s are optimized to `BroadcastedHashJoin` according to collected in-memory table statistics.
This PR fixes this issue by clearing the cache before testing join operator selection. A separate test case is also added to test broadcasted hash join operator selection.
Author: Cheng Lian <lian@databricks.com>
Closes #2960 from liancheng/fix-join-suite and squashes the following commits:
715b2de [Cheng Lian] Fixes caching related JoinSuite failure
Diffstat (limited to 'LICENSE')
0 files changed, 0 insertions, 0 deletions