[ https://issues.apache.org/jira/browse/HIVE-14731?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16220885#comment-16220885 ]
Zhiyuan Yang commented on HIVE-14731: ------------------------------------- [~kgyrtkirk] The main reason is mapjoin decide parallelism according to #splits, which is not good enough for cross product. The cost of xprod is mostly determined by #records instead of #split. > Use Tez cartesian product edge in Hive (unpartitioned case only) > ---------------------------------------------------------------- > > Key: HIVE-14731 > URL: https://issues.apache.org/jira/browse/HIVE-14731 > Project: Hive > Issue Type: Bug > Reporter: Zhiyuan Yang > Assignee: Zhiyuan Yang > Attachments: HIVE-14731.1.patch, HIVE-14731.10.patch, > HIVE-14731.11.patch, HIVE-14731.12.patch, HIVE-14731.13.patch, > HIVE-14731.14.patch, HIVE-14731.15.patch, HIVE-14731.16.patch, > HIVE-14731.17.patch, HIVE-14731.18.patch, HIVE-14731.19.patch, > HIVE-14731.2.patch, HIVE-14731.20.patch, HIVE-14731.21.patch, > HIVE-14731.22.patch, HIVE-14731.23.patch, HIVE-14731.3.patch, > HIVE-14731.4.patch, HIVE-14731.5.patch, HIVE-14731.6.patch, > HIVE-14731.7.patch, HIVE-14731.8.patch, HIVE-14731.9.patch, > HIVE-14731.addendum.patch > > > Given cartesian product edge is available in Tez now (see TEZ-3230), let's > integrate it into Hive on Tez. This allows us to have more than one reducer > in cross product queries. -- This message was sent by Atlassian JIRA (v6.4.14#64029)