Uploaded image for project: 'Phoenix Tephra'
  1. Phoenix Tephra
  2. TEPHRA-299

Executing a large batch delete is very slow

    XMLWordPrintableJSON

Details

    • Bug
    • Status: Closed
    • Major
    • Resolution: Fixed
    • 0.15.0-incubating
    • 0.16.0
    • None
    • None

    Description

      I noticed that batch deletes are quire slow. In the profiler I found that almost all of the time is spent in org.apache.hadoop.hbase.regionserver.wal.FSHLog.blockOnSync().

      Looking at TransactionProcessor.preDelete it is obvious why:

      The batch delete is translated into single puts that are added to the region one by one, so each time the WAL is flushed.

       

      Attachments

        1. 299-client.txt
          9 kB
          Lars Hofhansl
        2. 299-client-v2.txt
          9 kB
          Lars Hofhansl
        3. 299-client-v3.txt
          6 kB
          Lars Hofhansl
        4. 299-complete.txt
          17 kB
          Lars Hofhansl
        5. 299-DOES-NOT_WORK.txt
          6 kB
          Lars Hofhansl

        Issue Links

          Activity

            People

              larsh Lars Hofhansl
              larsh Lars Hofhansl
              Votes:
              0 Vote for this issue
              Watchers:
              5 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: