mapreduce/src/contrib/vaidya/src/java/org/apache/hadoop/vaidya/postexdiagnosis/tests/MapSideDiskSpill.java

/**
 * Licensed to the Apache Software Foundation (ASF) under one
 * or more contributor license agreements.  See the NOTICE file
 * distributed with this work for additional information
 * regarding copyright ownership.  The ASF licenses this file
 * to you under the Apache License, Version 2.0 (the
 * "License"); you may not use this file except in compliance
 * with the License.  You may obtain a copy of the License at
 *
 *     http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */
package org.apache.hadoop.vaidya.postexdiagnosis.tests;

import org.apache.hadoop.mapreduce.MRJobConfig;
import org.apache.hadoop.vaidya.statistics.job.JobStatistics;
import org.apache.hadoop.vaidya.statistics.job.JobStatisticsInterface.JobKeys;
import org.apache.hadoop.vaidya.statistics.job.JobStatisticsInterface.KeyDataType;
import org.apache.hadoop.vaidya.statistics.job.JobStatisticsInterface.MapTaskKeys;
import org.apache.hadoop.vaidya.statistics.job.JobStatisticsInterface.ReduceTaskKeys;
import org.apache.hadoop.vaidya.statistics.job.MapTaskStatistics;
import org.apache.hadoop.vaidya.DiagnosticTest;
import org.w3c.dom.Element;
import java.util.Hashtable;
import java.util.List;

/**
 *
 */
public class MapSideDiskSpill extends DiagnosticTest {

  private double _impact;
  private JobStatistics _job;
  private long _numLocalBytesWrittenByMaps;
  
  /**            
   * 
   */
  public MapSideDiskSpill() {
  }

  /* 
   *  
   */
  @Override
  public double evaluate(JobStatistics job) {
  
    /*
     * Set the this._job
     */
    this._job = job;
      
    /*
     * Read the Normalization Factor
     */
    double normF = getInputElementDoubleValue("NormalizationFactor", 3.0);
    
    /*
     * Get the sorted map task list by number MapTaskKeys.OUTPUT_BYTES
     */
    List<MapTaskStatistics> smTaskList = job.getMapTaskList(MapTaskKeys.FILE_BYTES_WRITTEN, KeyDataType.LONG);
    int size = smTaskList.size();
    long numLocalBytesWrittenByMaps = 0;
    for (int i=0; i<size; i++) {
      numLocalBytesWrittenByMaps += smTaskList.get(i).getLongValue(MapTaskKeys.FILE_BYTES_WRITTEN);
    }
    this._numLocalBytesWrittenByMaps = numLocalBytesWrittenByMaps;
    
    /*
     * Map only job vs. map reduce job
     * For MapReduce job MAP_OUTPUT_BYTES are normally written by maps on local disk, so they are subtracted
     * from the localBytesWrittenByMaps.
     */
    if (job.getLongValue(JobKeys.TOTAL_REDUCES) > 0) {
      this._impact = (this._numLocalBytesWrittenByMaps - job.getLongValue(JobKeys.MAP_OUTPUT_BYTES))/job.getLongValue(JobKeys.MAP_OUTPUT_BYTES);
    } else {
      this._impact = this._numLocalBytesWrittenByMaps/job.getLongValue(JobKeys.MAP_OUTPUT_BYTES);
    }
    
    if (this._impact > normF) {
      this._impact = 1.0;
    } else {
      this._impact = this._impact/normF;
    }
    
    return this._impact;
    
  }
  
  /* (non-Javadoc)
   * @see org.apache.hadoop.contrib.utils.perfadvisor.diagnostic_rules.DiagnosticRule#getAdvice()
   */
  @Override
  public String getPrescription() {
    return 
    "* Use combiner to lower the map output size.\n" +
      "* Increase map side sort buffer size (" + MRJobConfig.IO_SORT_MB + 
      ":" + this._job.getJobConf().getInt(MRJobConfig.IO_SORT_MB, 0) + ").\n" +
      ") if number of Map Output Records are large. \n" +
      "* Increase (" + MRJobConfig.MAP_SORT_SPILL_PERCENT + ":" + 
      this._job.getJobConf().getInt(MRJobConfig.MAP_SORT_SPILL_PERCENT, 0) + 
      "), default 0.80 i.e. 80% of sort buffer size. \n";
  }

  /* (non-Javadoc)
   * @see org.apache.hadoop.contrib.utils.perfadvisor.diagnostic_rules.DiagnosticRule#getReferenceDetails()
   */
  @Override
  public String getReferenceDetails() {
    String ref = 
    "* TotalMapOutputBytes: "+this._job.getLongValue(JobKeys.MAP_OUTPUT_BYTES)+"\n"+
    "* Total Local Bytes Written by Maps: "+this._numLocalBytesWrittenByMaps+"\n"+
    "* Impact: "+ truncate(this._impact);
    return ref;
  }
}