kgyrtkirk commented on a change in pull request #1347:
URL: https://github.com/apache/hive/pull/1347#discussion_r522870090



##########
File path: 
itests/qtest/src/test/java/org/apache/hadoop/hive/cli/TestTezTPCDS30TBPerfCliDriver.java
##########
@@ -17,30 +17,32 @@
  */
 package org.apache.hadoop.hive.cli;
 
-import java.io.File;
-import java.util.Comparator;
-import java.util.List;
-
 import org.apache.hadoop.hive.cli.control.CliAdapter;
 import org.apache.hadoop.hive.cli.control.CliConfigs;
+import org.apache.hadoop.hive.cli.control.SplitSupport;
 import org.junit.ClassRule;
 import org.junit.Rule;
 import org.junit.Test;
 import org.junit.rules.TestRule;
 import org.junit.runner.RunWith;
 import org.junit.runners.Parameterized;
 import org.junit.runners.Parameterized.Parameters;
+import org.junit.runners.model.Statement;
+
+import java.io.File;
+import java.util.Comparator;
+import java.util.List;
 
 @RunWith(Parameterized.class)
-public class TestTezPerfCliDriver {
+public class TestTezTPCDS30TBPerfCliDriver {
 
-  static CliAdapter adapter = new 
CliConfigs.TezPerfCliConfig(false).getCliAdapter();
+  static CliAdapter adapter = new 
CliConfigs.TezTPCDS30TBCliConfig().getCliAdapter();
 
   @Parameters(name = "{0}")
   public static List<Object[]> getParameters() throws Exception {
     List<Object[]> parameters = adapter.getParameters();
     parameters.sort(new C1());
-    return parameters;
+    return SplitSupport.process(parameters, 
TestTezTPCDS30TBPerfCliDriver.class, 10);

Review comment:
       20 minutes is also fine...we have quite a few [replication tests above 
30 minutes right 
now](http://ci.hive.apache.org/job/hive-precommit/job/master/337/testReport/org.apache.hadoop.hive.ql.parse/)
   
   
   note that in its current form it would not work because you need to have a 
separated "N_SPLITS" integer in the class source(like in other clidriver 
classes) for the splitter to work.
   
   I think we can leave this out for now - 10 splits is a bit too many for a 20 
minute test (2 or 3 would fit better)
   in case of this test there is a bit more overhead per execution; because it 
will probably need to download and launch the metastore.
   
   
   
   




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org



---------------------------------------------------------------------
To unsubscribe, e-mail: gitbox-unsubscr...@hive.apache.org
For additional commands, e-mail: gitbox-h...@hive.apache.org

Reply via email to