647b6f9315b871ff7078b9706adc94e06e38146e,perfkitbenchmarker/benchmarks/hadoop_terasort_benchmark.py,,Run,#Any#,149

Before Change

      1024 * 1024.0)
  master.RemoteCommand(hadoop_cmd + "teravalidate /terasort /teravalidate")

  master.RemoteCommand("hadoop-%s/bin/hadoop fs -rm -r /teragen" %
                       HADOOP_VERSION)
  master.RemoteCommand("hadoop-%s/bin/hadoop fs -rm -r /terasort" %
                       HADOOP_VERSION)
  master.RemoteCommand("hadoop-%s/bin/hadoop fs -rm -r /teravalidate" %
                       HADOOP_VERSION)

After Change

  vms = benchmark_spec.vms
  master = vms[0]

  hadoop_cmd = ("{0} jar {1}/share/hadoop/mapreduce/"
                "hadoop-mapreduce-examples-{2}.jar").format(
                    os.path.join(hadoop.HADOOP_BIN, "yarn"),
                    hadoop.HADOOP_DIR,
                    hadoop.HADOOP_VERSION)
  master.RemoteCommand("{0} teragen {1} /teragen".format(
      hadoop_cmd, FLAGS.terasort_num_rows))
  num_cpus = sum(vm.num_cpus for vm in vms[1:])
  start_time = time.time()

Instances