Uploaded image for project: 'Lucene - Core'
  1. Lucene - Core
  2. LUCENE-9378

Configurable compression for BinaryDocValues

    XMLWordPrintableJSON

Details

    • Improvement
    • Status: Resolved
    • Major
    • Resolution: Fixed
    • None
    • 8.8
    • None
    • None
    • New

    Description

      Lucene 8.5.1 includes a change to always compress BinaryDocValues. This caused (~30%) reduction in our red-line QPS (throughput). 

      We think users should be given some way to opt-in for this compression feature instead of always being enabled which can have a substantial query time cost as we saw during our upgrade. mikemccand suggested one possible approach by introducing a mode in Lucene80DocValuesFormat (COMPRESSED and UNCOMPRESSED) and allowing users to create a custom Codec subclassing the default Codec and pick the format they want.

      Idea is similar to Lucene50StoredFieldsFormat which has two modes, Mode.BEST_SPEED and Mode.BEST_COMPRESSION.

      Here's related issues for adding benchmark covering BINARY doc values query-time performance - https://github.com/mikemccand/luceneutil/issues/61

      Attachments

        1. hotspots-v76x.png
          129 kB
          Alex Klibisz
        2. hotspots-v76x.png
          129 kB
          Alex Klibisz
        3. hotspots-v76x.png
          129 kB
          Alex Klibisz
        4. hotspots-v76x.png
          129 kB
          Alex Klibisz
        5. hotspots-v76x.png
          129 kB
          Alex Klibisz
        6. hotspots-v77x.png
          136 kB
          Alex Klibisz
        7. hotspots-v77x.png
          136 kB
          Alex Klibisz
        8. hotspots-v77x.png
          136 kB
          Alex Klibisz
        9. hotspots-v77x.png
          136 kB
          Alex Klibisz
        10. image-2020-06-12-22-17-30-339.png
          14 kB
          Alex Klibisz
        11. image-2020-06-12-22-17-53-961.png
          14 kB
          Alex Klibisz
        12. image-2020-06-12-22-18-24-527.png
          14 kB
          Alex Klibisz
        13. image-2020-06-12-22-18-48-919.png
          14 kB
          Alex Klibisz
        14. snapshots-v76x.nps
          28 kB
          Alex Klibisz
        15. snapshots-v76x.nps
          28 kB
          Alex Klibisz
        16. snapshots-v76x.nps
          28 kB
          Alex Klibisz
        17. snapshot-v77x.nps
          32 kB
          Alex Klibisz
        18. snapshot-v77x.nps
          32 kB
          Alex Klibisz
        19. snapshot-v77x.nps
          32 kB
          Alex Klibisz

        Issue Links

          Activity

            People

              Unassigned Unassigned
              viral.gandhi Viral Gandhi
              Votes:
              4 Vote for this issue
              Watchers:
              12 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved:

                Time Tracking

                  Estimated:
                  Original Estimate - Not Specified
                  Not Specified
                  Remaining:
                  Remaining Estimate - 0h
                  0h
                  Logged:
                  Time Spent - 5h 50m
                  5h 50m