MAPREDUCE-5399. Unnecessary Configuration instantiation in IFileInputStream slows down merge. (Stanislav Barton via Sandy Ryza)

git-svn-id: https://svn.apache.org/repos/asf/hadoop/common/branches/branch-2.1.0-beta@1510818 13f79535-47bb-0310-9956-ffa450edef68

hadoop-mapreduce-project/CHANGES.txt

@@ -445,6 +445,9 @@ Release 2.1.0-beta - 2013-08-06
     MAPREDUCE-5419. TestSlive is getting FileNotFound Exception (Robert Parker
     via jlowe)
 
+    MAPREDUCE-5399. Unnecessary Configuration instantiation in IFileInputStream
+    slows down merge. (Stanislav Barton via Sandy Ryza)
+
   BREAKDOWN OF HADOOP-8562 SUBTASKS
 
     MAPREDUCE-4739. Some MapReduce tests fail to find winutils.

hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapred/BackupStore.java

@@ -81,6 +81,8 @@ public class BackupStore<K,V> {
   private boolean inReset = false;
   private boolean clearMarkFlag = false;
   private boolean lastSegmentEOF = false;
+  
+  private Configuration conf;
 
   public BackupStore(Configuration conf, TaskAttemptID taskid)
   throws IOException {
@@ -106,6 +108,8 @@ public class BackupStore<K,V> {
     fileCache = new FileCache(conf);
     tid = taskid;
     
+    this.conf = conf;
+    
     LOG.info("Created a new BackupStore with a memory of " + maxSize);
 
   }
@@ -500,7 +504,7 @@ public class BackupStore<K,V> {
       Reader<K, V> reader = 
         new org.apache.hadoop.mapreduce.task.reduce.InMemoryReader<K, V>(null, 
             (org.apache.hadoop.mapred.TaskAttemptID) tid, 
-            dataOut.getData(), 0, usedSize);
+            dataOut.getData(), 0, usedSize, conf);
       Segment<K, V> segment = new Segment<K, V>(reader, false);
       segmentList.add(segment);
       LOG.debug("Added Memory Segment to List. List Size is " + 

hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapreduce/task/reduce/InMemoryReader.java

@@ -24,6 +24,7 @@ import java.io.IOException;
 
 import org.apache.hadoop.classification.InterfaceAudience;
 import org.apache.hadoop.classification.InterfaceStability;
+import org.apache.hadoop.conf.Configuration;
 import org.apache.hadoop.io.DataInputBuffer;
 import org.apache.hadoop.mapred.IFile.Reader;
 import org.apache.hadoop.mapreduce.TaskAttemptID;
@@ -39,11 +40,11 @@ public class InMemoryReader<K, V> extends Reader<K, V> {
   DataInputBuffer memDataIn = new DataInputBuffer();
   private int start;
   private int length;
-
+  
   public InMemoryReader(MergeManagerImpl<K,V> merger, TaskAttemptID taskAttemptId,
-                        byte[] data, int start, int length)
+                        byte[] data, int start, int length, Configuration conf)
   throws IOException {
-    super(null, null, length - start, null, null);
+    super(conf, null, length - start, null, null);
     this.merger = merger;
     this.taskAttemptId = taskAttemptId;
 

hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/main/java/org/apache/hadoop/mapreduce/task/reduce/MergeManagerImpl.java

@@ -613,7 +613,7 @@ public class MergeManagerImpl<K, V> implements MergeManager<K, V> {
       fullSize -= size;
       Reader<K,V> reader = new InMemoryReader<K,V>(MergeManagerImpl.this, 
                                                    mo.getMapId(),
-                                                   data, 0, (int)size);
+                                                   data, 0, (int)size, jobConf);
       inMemorySegments.add(new Segment<K,V>(reader, true, 
                                             (mo.isPrimaryMapOutput() ? 
                                             mergedMapOutputsCounter : null)));