in Hadoop, the MapReduce job ID is in the format job_201412081211_0002. This indicates that the job is the second job (the job number starts at 0001) and the job starts on December 8, 2014 12:11. a task belongs to the job, replacing the job ID's "job" prefix with "task", followed by a suffix that indicates which job is in the middle of the task. For example: task_201412081211_0002_m_000003, which indicates that the task belongs to the third map task of the job_201412081211_0002 Job (000003).
because the map or reduce task in mapreduce might fail and the reason for Hadoop guessing, the map or reduce task may be executed multiple times, which is the task attempt. Its ID form is: attempt_201412081211_0002_m_000003_0, which means that the attempt belongs to the first attempt of the task_201412081211_0002_m_000003 task.
Job (Job), task, and task attempt in Hadoop
