Details

    • Type: New Feature New Feature
    • Status: Closed
    • Priority: Major Major
    • Resolution: Fixed
    • Affects Version/s: None
    • Fix Version/s: 0.21.0
    • Component/s: contrib/fair-share
    • Labels:
      None
    • Hadoop Flags:
      Reviewed
    • Release Note:
      Added support for preemption in the fair scheduler. The new configuration options for enabling this are described in the fair scheduler documentation.

      Description

      Task preemption is necessary in a multi-user Hadoop cluster for two reasons: users might submit long-running tasks by mistake (e.g. an infinite loop in a map program), or tasks may be long due to having to process large amounts of data. The Fair Scheduler (HADOOP-3746) has a concept of guaranteed capacity for certain queues, as well as a goal of providing good performance for interactive jobs on average through fair sharing. Therefore, it will support preempting under two conditions:
      1) A job isn't getting its guaranteed share of the cluster for at least T1 seconds.
      2) A job is getting significantly less than its fair share for T2 seconds (e.g. less than half its share).

      T1 will be chosen smaller than T2 (and will be configurable per queue) to meet guarantees quickly. T2 is meant as a last resort in case non-critical jobs in queues with no guaranteed capacity are being starved.

      When deciding which tasks to kill to make room for the job, we will use the following heuristics:

      • Look for tasks to kill only in jobs that have more than their fair share, ordering these by deficit (most overscheduled jobs first).
      • For maps: kill tasks that have run for the least amount of time (limiting wasted time).
      • For reduces: similar to maps, but give extra preference for reduces in the copy phase where there is not much map output per task (at Facebook, we have observed this to be the main time we need preemption - when a job has a long map phase and its reducers are mostly sitting idle and filling up slots).
      1. mapreduce-551-branch20.txt
        127 kB
        Todd Lipcon
      2. hadoop-4665-v7e.patch
        132 kB
        Matei Zaharia
      3. hadoop-4665-v7d.patch
        131 kB
        Matei Zaharia
      4. hadoop-4665-v7c.patch
        131 kB
        Matei Zaharia
      5. hadoop-4665-v7b.patch
        130 kB
        Matei Zaharia
      6. hadoop-4665-v7.patch
        130 kB
        Matei Zaharia
      7. hadoop-4665-v6.patch
        98 kB
        Matei Zaharia
      8. hadoop-4665-v5.patch
        98 kB
        Matei Zaharia
      9. hadoop-4665-v4.patch
        44 kB
        Matei Zaharia
      10. hadoop-4665-v3.patch
        44 kB
        Matei Zaharia
      11. hadoop-4665-v2.patch
        44 kB
        Matei Zaharia
      12. hadoop-4665-v1b.patch
        45 kB
        Matei Zaharia
      13. hadoop-4665-v1.patch
        45 kB
        Matei Zaharia
      14. fs-preemption-v0.patch
        57 kB
        Matei Zaharia
      15. fairshare-patches.tar.gz
        31 kB
        Todd Lipcon

        Issue Links

          Activity

            People

            • Assignee:
              Matei Zaharia
              Reporter:
              Matei Zaharia
            • Votes:
              2 Vote for this issue
              Watchers:
              22 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved:

                Development