Uploaded image for project: 'HBase'
  1. HBase
  2. HBASE-1938

Make in-memory table scanning faster

    XMLWordPrintableJSON

Details

    • Improvement
    • Status: Closed
    • Blocker
    • Resolution: Fixed
    • None
    • 0.90.4, 0.92.0
    • Performance
    • None
    • Reviewed

    Description

      This issue is about profiling hbase to see if I can make hbase scans run faster when all is up in memory. Talking to some users, they are seeing about 1/4 million rows a second. It should be able to go faster than this (Scanning an array of objects, they can do about 4-5x this).

      Attachments

        1. test.patch
          2 kB
          Michael Stack
        2. caching-keylength-in-kv.patch
          1 kB
          Michael Stack
        3. MemStoreScanPerformance.java
          2 kB
          Michael Stack
        4. MemStoreScanPerformance.java
          2 kB
          Nicolas Liochon
        5. 20110726_1938_KeyValueSkipListSet.patch
          2 kB
          Nicolas Liochon
        6. 20110726_1938_MemStore.patch
          6 kB
          Nicolas Liochon
        7. 20110726_1938_MemStoreScanPerformance.java
          3 kB
          Nicolas Liochon
        8. MemStoreScanPerformance.java
          3 kB
          Nicolas Liochon
        9. 20110802_MemStore.patch
          3 kB
          Nicolas Liochon

        Activity

          People

            nkeywal Nicolas Liochon
            stack Michael Stack
            Votes:
            0 Vote for this issue
            Watchers:
            15 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: