Uploaded image for project: 'Core Server'
  1. Core Server
  2. SERVER-17382

rc10/wiredTiger multi collection/DB bulk insert slow than rc8 in initial insertion phase

    Details

    • Type: Bug
    • Status: Closed
    • Priority: Critical - P2
    • Resolution: Fixed
    • Affects Version/s: 3.0.0-rc10
    • Fix Version/s: 3.0.2, 3.1.1
    • Component/s: WiredTiger
    • Labels:
    • Backwards Compatibility:
      Fully Compatible
    • Operating System:
      ALL
    • Backport Completed:

      Description

      re-test rc10 for SERVER-17344, see improvement with insert throughput, but overall still slower than that of RC8.

      version insert throughput
      rc8 94,069.51
      rc9 57,129.39
      rc10 66,562.25

      after debug, found out that rc10

      • have much higher percentage of very slow insert (> 1 sec), see the attached chart from mtools.
      • is very slow for the first few minutes, it recovers and stabilizes after few minutes.

      Combine this with the way sysbench insert is designed (pin one thread one collection), it could amplify the impact in overall throughput.

      Run bulk insert with benchRun show insert throughput is similar between rc10 & rc8.

      I think the high number of very slow insert is still an issue.

        Attachments

        1. flamegraph_batch_insert.png
          flamegraph_batch_insert.png
          169 kB
        2. insert.png
          insert.png
          66 kB
        3. long-run-rc10_benchRun.log.html
          395 kB
        4. mongod-rc10-8col.log.html
          72 kB
        5. mongod-rc8-8col.log.html
          107 kB

          Issue Links

            Activity

              People

              • Votes:
                0 Vote for this issue
                Watchers:
                12 Start watching this issue

                Dates

                • Created:
                  Updated:
                  Resolved: