svn commit: r1467000 - in /hive/trunk: common/src/java/org/apache/hadoop/hive/conf/HiveConf.java conf/hive-default.xml.template ql/src/java/org/apache/hadoop/hive/ql/exec/CommonJoinOperator.java

[na...@apache.org]

Author: namit
Date: Thu Apr 11 17:46:47 2013
New Revision: 1467000

URL: http://svn.apache.org/r1467000
Log:
HIVE-4328 Change default bigtable selection policy for sort-merge joins
(Namit Jain via Ashutosh Chauhan)


Modified:
    hive/trunk/common/src/java/org/apache/hadoop/hive/conf/HiveConf.java
    hive/trunk/conf/hive-default.xml.template
    hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/exec/CommonJoinOperator.java

Modified: hive/trunk/common/src/java/org/apache/hadoop/hive/conf/HiveConf.java
URL: http://svn.apache.org/viewvc/hive/trunk/common/src/java/org/apache/hadoop/hive/conf/HiveConf.java?rev=1467000&r1=1466999&r2=1467000&view=diff
==============================================================================
--- hive/trunk/common/src/java/org/apache/hadoop/hive/conf/HiveConf.java (original)
+++ hive/trunk/common/src/java/org/apache/hadoop/hive/conf/HiveConf.java Thu Apr 11 17:46:47 2013
@@ -523,7 +523,7 @@ public class HiveConf extends Configurat
     HIVE_AUTO_SORTMERGE_JOIN("hive.auto.convert.sortmerge.join", false),
     HIVE_AUTO_SORTMERGE_JOIN_BIGTABLE_SELECTOR(
         "hive.auto.convert.sortmerge.join.bigtable.selection.policy",
-        "org.apache.hadoop.hive.ql.optimizer.LeftmostBigTableSelectorForAutoSMJ"),
+        "org.apache.hadoop.hive.ql.optimizer.AvgPartitionSizeBasedBigTableSelectorForAutoSMJ"),
 
     HIVESCRIPTOPERATORTRUST("hive.exec.script.trust", false),
     HIVEROWOFFSET("hive.exec.rowoffset", false),

Modified: hive/trunk/conf/hive-default.xml.template
URL: http://svn.apache.org/viewvc/hive/trunk/conf/hive-default.xml.template?rev=1467000&r1=1466999&r2=1467000&view=diff
==============================================================================
--- hive/trunk/conf/hive-default.xml.template (original)
+++ hive/trunk/conf/hive-default.xml.template Thu Apr 11 17:46:47 2013
@@ -969,9 +969,11 @@
 
 <property>
   <name>hive.auto.convert.sortmerge.join.bigtable.selection.policy</name>
-  <value>org.apache.hadoop.hive.ql.optimizer.LeftmostBigTableSelectorForAutoSMJ</value>
+  <value>org.apache.hadoop.hive.ql.optimizer.AvgPartitionSizeBasedBigTableSelectorForAutoSMJ</value>
   <description>The policy to choose the big table for automatic conversion to sort-merge join.
-    By default, the leftmost table is assigned the big table. Other policies are based on size:
+    By default, the table with the largest partitions is assigned the big table. All policies are:
+    . based on position of the table - the leftmost table is selected
+    org.apache.hadoop.hive.ql.optimizer.LeftmostBigTableSMJ.
     . based on total size (all the partitions selected in the query) of the table 
     org.apache.hadoop.hive.ql.optimizer.TableSizeBasedBigTableSelectorForAutoSMJ.
     . based on average size (all the partitions selected in the query) of the table 

Modified: hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/exec/CommonJoinOperator.java
URL: http://svn.apache.org/viewvc/hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/exec/CommonJoinOperator.java?rev=1467000&r1=1466999&r2=1467000&view=diff
==============================================================================
--- hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/exec/CommonJoinOperator.java (original)
+++ hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/exec/CommonJoinOperator.java Thu Apr 11 17:46:47 2013
@@ -385,7 +385,7 @@ public abstract class CommonJoinOperator
   // all evaluation should be processed here for valid aliasFilterTags
   //
   // for MapJoin, filter tag is pre-calculated in MapredLocalTask and stored with value.
-  // when reading the hashtable, MapJoinObjectValue calcuates alias filter and provide it to join
+  // when reading the hashtable, MapJoinObjectValue calculates alias filter and provide it to join
   protected ArrayList<Object> getFilteredValue(byte alias, Object row) throws HiveException {
     boolean hasFilter = hasFilter(alias);
     ArrayList<Object> nr = JoinUtil.computeValues(row, joinValues[alias],