----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/27720/#review60316 -----------------------------------------------------------
1. Looking at the diff, I was sure where we are removing unnecessary counter registrations. 2. It would be great if we can have some tests that are enabled with counter statistics collection, so that we know what kind of output we are expecting and avoid future breakage. - Xuefu Zhang On Nov. 7, 2014, 5:27 a.m., chengxiang li wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/27720/ > ----------------------------------------------------------- > > (Updated Nov. 7, 2014, 5:27 a.m.) > > > Review request for hive and Xuefu Zhang. > > > Bugs: HIVE-8777 > https://issues.apache.org/jira/browse/HIVE-8777 > > > Repository: hive-git > > > Description > ------- > > Currently we register all hive operator counters in SparkCounters, while > actually not all hive operators are used in SparkTask, we should iterate > SparkTask's operators, and only register conuters required. > > > Diffs > ----- > > ql/src/java/org/apache/hadoop/hive/ql/exec/spark/SparkClient.java e955da3 > ql/src/java/org/apache/hadoop/hive/ql/exec/spark/SparkTask.java 46b04bc > ql/src/java/org/apache/hadoop/hive/ql/exec/spark/counter/SparkCounters.java > bb3597a > ql/src/java/org/apache/hadoop/hive/ql/plan/SparkWork.java 66fd6b6 > > Diff: https://reviews.apache.org/r/27720/diff/ > > > Testing > ------- > > > Thanks, > > chengxiang li > >
