[ https://issues.apache.org/jira/browse/SPARK-17050?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15420336#comment-15420336 ]
Sean Owen commented on SPARK-17050: ----------------------------------- [~WeichenXu123] this sounds quite related to https://issues.apache.org/jira/browse/SPARK-17033 . I don't think we should open JIRAs for the exact same logical change in different classes. Let's put these together into SPARK-17033. > Improve initKMeansParallel with treeAggregate > --------------------------------------------- > > Key: SPARK-17050 > URL: https://issues.apache.org/jira/browse/SPARK-17050 > Project: Spark > Issue Type: Improvement > Reporter: Weichen Xu > Original Estimate: 24h > Remaining Estimate: 24h > > The `initKMeansParallel` use `rdd.aggregate`, it is better to use > `treeAggregate` to get better performance. -- This message was sent by Atlassian JIRA (v6.3.4#6332) --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org