Uploaded image for project: 'Hadoop Map/Reduce'
  1. Hadoop Map/Reduce
  2. MAPREDUCE-5962

Support CRC32C in IFile

    XMLWordPrintableJSON

Details

    • Improvement
    • Status: Resolved
    • Major
    • Resolution: Won't Fix
    • 2.5.0
    • None
    • performance, task
    • None

    Description

      Currently, the IFile format used by the MR shuffle checksums all data using the zlib CRC32 polynomial. If we allow use of CRC32C instead, we can get a large reduction in CPU usage by leveraging the native hardware CRC32C implementation (approx half a second of CPU time savings per GB checksummed).

      Attachments

        1. mapreduce-5962.txt
          14 kB
          Todd Lipcon
        2. mapreduce-5962.txt
          14 kB
          Todd Lipcon

        Issue Links

          is related to

          Improvement - An improvement or enhancement to an existing feature or task. HDFS-3528 Use native CRC32 in DFS write path

          • Major - Major loss of function.
          • Closed
          relates to

          Improvement - An improvement or enhancement to an existing feature or task. MAPREDUCE-2841 Task level native optimization

          • Major - Major loss of function.
          • Resolved

          Improvement - An improvement or enhancement to an existing feature or task. HADOOP-10859 Native implementation of java Checksum interface

          • Minor - Minor loss of function, or other problem where easy workaround is present.
          • Resolved

          Activity

            People

              tlipcon Todd Lipcon
              tlipcon Todd Lipcon
              Votes:
              0 Vote for this issue
              Watchers:
              8 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: