Uploaded image for project: 'Red Hat OpenShift Data Science'
  1. Red Hat OpenShift Data Science
  2. RHODS-5190

Average spawn time is 50 seconds

XMLWordPrintable

    • False
    • None
    • False
    • Yes
    • Yes
    • No
    • Yes
    • None

      Description of problem:

      The average spawn time (measured from clicking the spawn button to the user being in the JL UI) for notebook images in KFNBC is around 50 seconds, which is 10 seconds slower than in JH.
      I've tested this twice with 10 spawns each time, getting results slightly below or slightly over 50s. This was done on a "default" size OSD cluster.
      This is a regression compared to the earlier performance and breaks one of the requirements laid out by jdemoss@redhat.com 

      Prerequisites (if any, like setup, operators/versions):

      Latest RHODS 1.16 RC

      Steps to Reproduce

      1. Go to spawner page
      2. spawn image and start timing
      3. wait until the JL UI is fully loaded and stop timing
      4. Average over multiple spawns (e.g. 10)

      Actual results:

      Avg. spawn time is ~50s

      Expected results:

      Avg. spawn time is <=40s

      Reproducibility (Always/Intermittent/Only Once):

      always

      Build Details:

      Workaround:

      Additional info:

        1. avg_time
          296 kB
        2. avg_time.png
          avg_time.png
          296 kB
        3. avg_time2.png
          avg_time2.png
          79 kB
        4. image-2022-09-12-11-39-18-438.png
          image-2022-09-12-11-39-18-438.png
          296 kB
        5. image-2022-09-12-11-42-02-714.png
          image-2022-09-12-11-42-02-714.png
          91 kB
        6. image-2022-10-25-17-00-17-091.png
          image-2022-10-25-17-00-17-091.png
          19 kB

            lferrnan@redhat.com Lucas Fernandez Aragon
            rhn-support-lgiorgi Luca Giorgi
            Votes:
            0 Vote for this issue
            Watchers:
            6 Start watching this issue

              Created:
              Updated:
              Resolved: