Uploaded image for project: 'Apache YuniKorn'
  1. Apache YuniKorn
  2. YUNIKORN-85

Improve recovery performance by querying all pods once instead of querying pods on specified node for many times

    XMLWordPrintableJSON

    Details

      Description

      Recently I did some performance tests on kubemark cluster with 1k hollow nodes, and found the recovery phase will take almost 1 minute, querying pods on specified node one by one is the most time-consuming logic inside. Therefore I propose to query all pods once here to improve the recovery performance, after this improvement, the cost time of recovery phase can reduce to only 2 seconds.

        Attachments

          Issue Links

            Activity

              People

              • Assignee:
                taoyang Tao Yang
                Reporter:
                Tao Yang Tao Yang
              • Votes:
                0 Vote for this issue
                Watchers:
                2 Start watching this issue

                Dates

                • Created:
                  Updated:
                  Resolved:

                  Time Tracking

                  Estimated:
                  Original Estimate - Not Specified
                  Not Specified
                  Remaining:
                  Remaining Estimate - 0h
                  0h
                  Logged:
                  Time Spent - 20m
                  20m