Uploaded image for project: 'Hive'
  1. Hive
  2. HIVE-6430

MapJoin hash table has large memory overhead

    Details

    • Type: Improvement
    • Status: Closed
    • Priority: Major
    • Resolution: Fixed
    • Affects Version/s: None
    • Fix Version/s: 0.14.0
    • Component/s: None
    • Labels:
      None

      Description

      Right now, in some queries, I see that storing e.g. 4 ints (2 for key and 2 for row) can take several hundred bytes, which is ridiculous. I am reducing the size of MJKey and MJRowContainer in other jiras, but in general we don't need to have java hash table there. We can either use primitive-friendly hashtable like the one from HPPC (Apache-licenced), or some variation, to map primitive keys to single row storage structure without an object per row (similar to vectorization).

        Attachments

        1. HIVE-6430.patch
          134 kB
          Sergey Shelukhin
        2. HIVE-6430.01.patch
          149 kB
          Sergey Shelukhin
        3. HIVE-6430.02.patch
          137 kB
          Sergey Shelukhin
        4. HIVE-6430.03.patch
          149 kB
          Sergey Shelukhin
        5. HIVE-6430.04.patch
          158 kB
          Sergey Shelukhin
        6. HIVE-6430.05.patch
          162 kB
          Sergey Shelukhin
        7. HIVE-6430.06.patch
          161 kB
          Sergey Shelukhin
        8. HIVE-6430.07.patch
          169 kB
          Sergey Shelukhin
        9. HIVE-6430.08.patch
          170 kB
          Sergey Shelukhin
        10. HIVE-6430.09.patch
          179 kB
          Sergey Shelukhin
        11. HIVE-6430.10.patch
          195 kB
          Sergey Shelukhin
        12. HIVE-6430.11.patch
          202 kB
          Sergey Shelukhin
        13. HIVE-6430.12.patch
          204 kB
          Sergey Shelukhin
        14. HIVE-6430.12.patch
          204 kB
          Sergey Shelukhin
        15. HIVE-6430.13.patch
          205 kB
          Sergey Shelukhin
        16. HIVE-6430.14.patch
          207 kB
          Sergey Shelukhin

          Issue Links

            Activity

              People

              • Assignee:
                sershe Sergey Shelukhin
                Reporter:
                sershe Sergey Shelukhin
              • Votes:
                0 Vote for this issue
                Watchers:
                8 Start watching this issue

                Dates

                • Created:
                  Updated:
                  Resolved: