Uploaded image for project: 'Spark'
  1. Spark
  2. SPARK-36735

Adjust overhead of cached relation for DPP

    XMLWordPrintableJSON

Details

    • Improvement
    • Status: Resolved
    • Major
    • Resolution: Fixed
    • 3.3.0
    • 3.3.0
    • SQL
    • None

    Description

      Currently we calculate if there is benefit of pruning with DPP by simply summing up the size of all scan relations as the overhead. However, for cached relations, the overhead should be different than a non-cached relation. This proposes to use adjusted overhead for cached relation with DPP.

      Attachments

        Activity

          People

            viirya L. C. Hsieh
            viirya L. C. Hsieh
            Votes:
            0 Vote for this issue
            Watchers:
            2 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: