Uploaded image for project: 'IMPALA'
  1. IMPALA
  2. IMPALA-6311

Evaluate smaller FPP for Bloom filters

    XMLWordPrintableJSON

Details

    • Task
    • Status: Resolved
    • Major
    • Resolution: Fixed
    • None
    • Impala 4.5.0
    • Perf Investigation
    • None

    Description

      The Bloom filters are created by estimating the NDV and then using the FPP of 75% to get the right size for the filter. This is may be too high to be very useful - if our filters are currently filtering more than 75% out, then it is only because we are overestimating NDV.

      Attachments

        Issue Links

          depends upon

          Improvement - An improvement or enhancement to an existing feature or task. IMPALA-10110 Separate option to control fpp for bloom filter sizing

          • Major - Major loss of function.
          • Resolved
          is related to

          Bug - A problem which impairs or prevents the functions of the product. IMPALA-5633 Bloom filters underestimate false positive probability

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

          Bug - A problem which impairs or prevents the functions of the product. IMPALA-11924 Bloom filter size is unaffected by column NDV

          • Major - Major loss of function.
          • Resolved

          Improvement - An improvement or enhancement to an existing feature or task. IMPALA-12451 Cardinality underestimation can hurt bloom filter effectiveness

          • Major - Major loss of function.
          • Open

          Activity

            People

              rizaon Riza Suminto
              jbapple Jim Apple
              Votes:
              0 Vote for this issue
              Watchers:
              8 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: