Hadoop多Job并行处理的实例详解

<property>
 <name>mapred.jobtracker.taskScheduler</name>
 <value>org.apache.hadoop.mapred.FairScheduler</value>
</property>

<property>
 <name>hadoopTest.jar</name>
 <value>所生成jar的地址</value>
</property>

// 获取各个Job，有关Job的创建，这里就不贴上来了。
Job job_base = (Job) 。。。;
Job job_avg = (Job) 。。。;
Job job_runCount = (Job) 。。。;
Job job_activeUser = (Job) ....;


job_base.setJarByClass(CapuseDateTimerTask.class);
job_avg.setJarByClass(CapuseDateTimerTask.class);
job_runCount.setJarByClass(CapuseDateTimerTask.class);
job_activeUser.setJarByClass(CapuseDateTimerTask.class);


// 执行完job_base才开始并行以下三个Job；
if (job_base.waitForCompletion(true)) {
FileUtil.hdfsFileHandle(jobBase);


// 并行Job
job_avg.submit();
job_runCount.submit();
job_activeUser.submit();
}

boolean bln1 = job_avg.isComplete();
boolean bln2 = job_runCount.isComplete();
boolean bln3 = job_activeUser.isComplete();


// 计算Job是否完成
while (!bln1 || !bln2 || !bln3) {
bln1 = job_avg.isComplete();
bln2 = job_runCount.isComplete();
bln3 = job_activeUser.isComplete();
}

hadoop jar hadoopTest.jar ch03.test Test