diff options
author | Mridul Muralidharan <mridulm@yahoo-inc.com> | 2015-05-01 21:23:42 -0700 |
---|---|---|
committer | Reynold Xin <rxin@databricks.com> | 2015-05-01 21:23:42 -0700 |
commit | b79aeb95b45ab4ae811039d452cf028d7b844132 (patch) | |
tree | 4d6215b96ab2e8f3848045c1d9ca1b8ebded6011 /python/pyspark/sql/tests.py | |
parent | 38d4e9e446b425ca6a8fe8d8080f387b08683842 (diff) | |
download | spark-b79aeb95b45ab4ae811039d452cf028d7b844132.tar.gz spark-b79aeb95b45ab4ae811039d452cf028d7b844132.tar.bz2 spark-b79aeb95b45ab4ae811039d452cf028d7b844132.zip |
[SPARK-7317] [Shuffle] Expose shuffle handle
Details in JIRA, in a nut-shell, all machinary for custom RDD's to leverage spark shuffle directly (without exposing impl details of shuffle) exists - except for this small piece.
Exposing this will allow for custom dependencies to get a handle to ShuffleHandle - which they can then leverage on reduce side.
Author: Mridul Muralidharan <mridulm@yahoo-inc.com>
Closes #5857 from mridulm/expose_shuffle_handle and squashes the following commits:
d8b6bd4 [Mridul Muralidharan] Expose ShuffleHandle
Diffstat (limited to 'python/pyspark/sql/tests.py')
0 files changed, 0 insertions, 0 deletions