Uploaded image for project: 'Hadoop Map/Reduce'
  1. Hadoop Map/Reduce
  2. MAPREDUCE-5268

Improve history server startup performance

    XMLWordPrintableJSON

Details

    • Improvement
    • Status: Closed
    • Major
    • Resolution: Fixed
    • 0.23.7, 2.0.4-alpha
    • 2.1.0-beta, 0.23.9
    • jobhistoryserver
    • None
    • Reviewed

    Description

      The history server can easily take many minutes to startup when there are a significant number of jobs to scan in the done directory. However the scanning of files is not the bottleneck, rather it's the heavy use of ConcurrentSkipListMap.size in HistoryFileManager.

      ConcurrentSkipListMap.size is a very expensive operation, especially on maps with many entries, as it has to scan every entry to compute the size. We should avoid calling this method or at least minimize its use.

      Attachments

        1. mr-5268-prelim.patch
          5 kB
          Karthik Kambatla
        2. mr-5268.patch
          9 kB
          Karthik Kambatla
        3. mr-5268.patch
          10 kB
          Karthik Kambatla
        4. mr-5268.patch
          11 kB
          Karthik Kambatla
        5. mr-5268.patch
          11 kB
          Karthik Kambatla
        6. mr-5268.patch
          10 kB
          Karthik Kambatla

        Activity

          People

            kasha Karthik Kambatla
            jlowe Jason Darrell Lowe
            Votes:
            0 Vote for this issue
            Watchers:
            8 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: