Uploaded image for project: 'Hive'
  1. Hive
  2. HIVE-21548

Statistics: String statistics computation is slower with vectorization enabled

    XMLWordPrintableJSON

Details

    • Improvement
    • Status: Open
    • Major
    • Resolution: Unresolved
    • 4.0.0
    • None
    • Statistics, Vectorization
    • None

    Description

      Computing the hashCode of a String for HLL is going

      byte[] -> Text -> String -> byte[]

      to get the Murmur3 hashCode of the byte[] region.

      Attachments

        1. string-stats-hll.png
          70 kB
          Gopal Vijayaraghavan
        2. prof-hll-string-stats.svg
          114 kB
          Gopal Vijayaraghavan

        Activity

          People

            Unassigned Unassigned
            gopalv Gopal Vijayaraghavan
            Votes:
            0 Vote for this issue
            Watchers:
            2 Start watching this issue

            Dates

              Created:
              Updated: