Uploaded image for project: 'WiredTiger'
  1. WiredTiger
  2. WT-6164

Reduce the number of page flags according to the flag variable size

XMLWordPrintableJSON

    • Type: Icon: Build Failure Build Failure
    • Resolution: Fixed
    • Priority: Icon: Major - P3 Major - P3
    • WT10.0.0, 4.4.0-rc6, 4.7.0
    • Affects Version/s: None
    • Component/s: None
    • None
    • Storage - Ra 2020-05-18

      It has been observed that MongoDB is using a lot of memory during the shutdown phase following some benchmark runs. We should investigate whether rollback_to_stable is the cause of those memory allocations, and figure out what we can do to alleviate the pressure.

        1. b1.tar.gz
          2.63 MB
          Mark Callaghan
        2. comparison_between_before_and_after_rc4.png
          32 kB
          Haribabu Kommi
        3. Linkbench2-profile.png
          151 kB
          David Daly
        4. Memory_increase_from_local_run.png
          16 kB
          Haribabu Kommi
        5. metrics.2020-05-09T23-31-24Z-00000.2g.oom
          10.01 MB
          Mark Callaghan
        6. metrics.2020-05-10T00-51-52Z-00000.2g.oom
          9.99 MB
          Mark Callaghan
        7. metrics.2020-05-10T02-11-08Z-00000.2g.oom
          10.01 MB
          Mark Callaghan
        8. metrics.2020-05-10T03-38-14Z-00000.2g.oom
          10.01 MB
          Mark Callaghan
        9. metrics.2020-05-10T05-03-20Z-00000.2g.oom
          9.99 MB
          Mark Callaghan
        10. metrics.2020-05-10T06-31-30Z-00000.2g.oom
          1.24 MB
          Mark Callaghan
        11. mongod.log.2g.oom.gz
          18.72 MB
          Mark Callaghan
        12. WT_shutdown_memory_increase.png
          72 kB
          Haribabu Kommi

            Assignee:
            haribabu.kommi@mongodb.com Haribabu Kommi
            Reporter:
            alexander.gorrod@mongodb.com Alexander Gorrod
            Votes:
            0 Vote for this issue
            Watchers:
            14 Start watching this issue

              Created:
              Updated:
              Resolved: