Uploaded image for project: 'Hive'
  1. Hive
  2. HIVE-18908

FULL OUTER JOIN to MapJoin

    XMLWordPrintableJSON

Details

    • Improvement
    • Status: Closed
    • Critical
    • Resolution: Fixed
    • None
    • 4.0.0-alpha-1
    • Hive
    • None

    Description

      Currently, we do not support FULL OUTER JOIN in MapJoin.

      Rough TPC-DS timings run on laptop:

      (NOTE: Query 51 has PTF as a bigger serial portion – Amdahl's law at play)

      FULL OUTER MapJoin OFF = MergeJoin

      Query 51:
      o Vectorization OFF
      • FULL OUTER MapJoin OFF: 4:30 minutes
      • FULL OUTER MapJoin ON: 4:37 minutes
      o Vectorization ON
      • FULL OUTER MapJoin OFF: 2:35 minutes
      • FULL OUTER MapJoin ON: 1:47 minutes

      Query 97:
      o Vectorization OFF
      • FULL OUTER MapJoin OFF: 2:37 minutes
      • FULL OUTER MapJoin ON: 2:42 minutes
      o Vectorization ON
      • FULL OUTER MapJoin OFF: 1:17 minutes
      • FULL OUTER MapJoin ON: 0:06 minutes

      FULL OUTER Join 10,000,000 rows against 323,910 small table keys
      o Vectorization ON
      • FULL OUTER MapJoin OFF: 14:56 minutes
      • FULL OUTER MapJoin ON: 1:45 minutes

      FULL OUTER Join 10,000,000 rows against 1,000 small table keys
      o Vectorization ON
      • FULL OUTER MapJoin OFF: 12:37 minutes
      • FULL OUTER MapJoin ON: 1:38 minutes

      Hopefully, someone will do large scale cluster testing. [DynamicPartitionedHashJoin] MapJoin should scale dramatically better than [Sort] MergeJoin reduce-shuffle.

      Attachments

        1. HIVE-18908.09998.patch
          4.64 MB
          Matt McCline
        2. HIVE-18908.09997.patch
          4.64 MB
          Matt McCline
        3. HIVE-18908.09996.patch
          4.64 MB
          Matt McCline
        4. HIVE-18908.09995.patch
          4.64 MB
          Matt McCline
        5. HIVE-18908.09994.patch
          4.64 MB
          Matt McCline
        6. HIVE-18908.09993.patch
          4.63 MB
          Matt McCline
        7. HIVE-18908.09992.patch
          4.63 MB
          Matt McCline
        8. HIVE-18908.09991.patch
          4.64 MB
          Matt McCline
        9. HIVE-18908.0999.patch
          4.64 MB
          Matt McCline
        10. HIVE-18908.0998.patch
          4.64 MB
          Matt McCline
        11. HIVE-18908.0997.patch
          4.57 MB
          Matt McCline
        12. HIVE-18908.0996.patch
          4.60 MB
          Matt McCline
        13. HIVE-18908.0995.patch
          3.52 MB
          Matt McCline
        14. HIVE-18908.0994.patch
          3.51 MB
          Matt McCline
        15. HIVE-18908.0993.patch
          3.51 MB
          Matt McCline
        16. HIVE-18908.0992.patch
          3.42 MB
          Matt McCline
        17. HIVE-18908.0991.patch
          3.42 MB
          Matt McCline
        18. HIVE-18908.099.patch
          3.42 MB
          Matt McCline
        19. HIVE-18908.098.patch
          3.38 MB
          Matt McCline
        20. HIVE-18908.097.patch
          3.39 MB
          Matt McCline
        21. HIVE-18908.096.patch
          3.03 MB
          Matt McCline
        22. HIVE-18908.093.patch
          5.86 MB
          Matt McCline
        23. HIVE-18908.092.patch
          5.75 MB
          Matt McCline
        24. HIVE-18908.091.patch
          5.79 MB
          Matt McCline
        25. HIVE-18908.09.patch
          5.79 MB
          Matt McCline
        26. HIVE-18908.08.patch
          6.01 MB
          Matt McCline
        27. FULL OUTER MapJoin Code Changes.docx
          129 kB
          Matt McCline
        28. HIVE-18908.06.patch
          5.08 MB
          Matt McCline
        29. JOIN to MAPJOIN Transformation.pdf
          26 kB
          Matt McCline
        30. SHARED-MEMORY FULL OUTER MapJoin.pdf
          58 kB
          Matt McCline
        31. HIVE-18908.05.patch
          4.76 MB
          Matt McCline
        32. HIVE-18908.04.patch
          4.78 MB
          Matt McCline
        33. HIVE-18908.03.patch
          4.85 MB
          Matt McCline
        34. HIVE-18908.02.patch
          4.84 MB
          Matt McCline
        35. HIVE-18908.01.patch
          2.19 MB
          Matt McCline

        Issue Links

          Activity

            People

              mmccline Matt McCline
              mmccline Matt McCline
              Votes:
              0 Vote for this issue
              Watchers:
              2 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: