Uploaded image for project: 'Spark'
  1. Spark
  2. SPARK-3239

Choose disks for spilling randomly in PySpark

    XMLWordPrintableJSON

    Details

    • Type: Improvement
    • Status: Resolved
    • Priority: Major
    • Resolution: Fixed
    • Affects Version/s: 1.1.0
    • Fix Version/s: 1.1.0
    • Component/s: PySpark
    • Labels:
      None
    • Target Version/s:

      Description

      Currently, ExternalHashMerger spill data into disks in the same order crossing processes in same node, it will have IO contention.

      It should choose the disks randomly.

        Attachments

          Activity

            People

            • Assignee:
              davies Davies Liu
              Reporter:
              davies Davies Liu
            • Votes:
              0 Vote for this issue
              Watchers:
              2 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved: