Uploaded image for project: 'Hadoop Map/Reduce'
  1. Hadoop Map/Reduce
  2. MAPREDUCE-4587

Support fetch by key boundaries for memcmp types

    XMLWordPrintableJSON

Details

    • Improvement
    • Status: Open
    • Major
    • Resolution: Unresolved
    • None
    • None
    • nodemanager, task
    • None

    Description

      Intermediate data addressable by key support not only restartable streams, but partitioning after the map output are written. With sampling of map output, a job can implement a total-order and tune the number of reduces around skew. It is possible to implement something similar for non-memcmp types, but it is significantly more complex.

      Attachments

        Issue Links

          Activity

            People

              Unassigned Unassigned
              cdouglas Christopher Douglas
              Votes:
              0 Vote for this issue
              Watchers:
              3 Start watching this issue

              Dates

                Created:
                Updated: