Uploaded image for project: 'Hadoop Map/Reduce'
  1. Hadoop Map/Reduce
  2. MAPREDUCE-6678

Allow ShuffleHandler readahead without drop-behind

    XMLWordPrintableJSON

Details

    • Improvement
    • Status: Resolved
    • Major
    • Resolution: Fixed
    • 2.7.2, 3.0.0-alpha1
    • 2.8.0, 3.0.0-alpha1
    • nodemanager
    • None

    Description

      Currently mapreduce.shuffle.manage.os.cache enables/disables both readahead (POSIX_FADV_WILLNEED) and drop-behind (POSIX_FADV_DONTNEED) logic within the ShuffleHandler.

      It would be beneficial if these were separately configurable.

      • Running without readahead can lead to significant seek storms caused by large numbers of sendfiles() competing with one another.
      • However, running with drop-behind can also lead to seek storms because there are cases where the server can successfully write the shuffle bytes to the network, BUT the client doesn't want the bytes right now (MergeManager wants to WAIT is an example) so it ignores them and asks for them again a bit later. This causes repeated reads of the same data from disk.

      I'll attach a simple patch that enables/disables readahead based on mapreduce.shuffle.readahead.bytes==0, leaving mapreduce.shuffle.manage.os.cache controlling only the drop-behind.

      Attachments

        1. YARN-4964.001.patch
          1 kB
          Nathan Roberts

        Activity

          People

            nroberts Nathan Roberts
            nroberts Nathan Roberts
            Votes:
            0 Vote for this issue
            Watchers:
            9 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: