Uploaded image for project: 'IGB'
  1. IGB
  2. IGBF-3631

Investigate: Responsiveness in IGB 10 versus IGB 9

    Details

    • Type: Task
    • Status: Closed (View Workflow)
    • Priority: Major
    • Resolution: Done
    • Affects Version/s: None
    • Fix Version/s: None
    • Labels:
      None

      Description

      To test:

      • Open IGB
      • Select latest human genome (click Mona Lisa icon); chromosome 1 will be the default chromosome upon selected the genome
      • Open RNA-Seq > Graph - Scaled
      • Click all the checkboxes to "open" all the datasets
      • Click Load Data
      • Wait for all the data to load from the server into IGB
      • Click the trashcan icon (lower right) to force garbage collection
      • Once all the data are loaded, choose the Graph tab
      • Within the Graph tab, click "Select All" to select all the graphs
      • Use the Y Axis Scale section to set Min value to 0 and Max value to 10000

      Compare IGB running in JDK21 (version 10) to IGB running in JDK8 (version 9)

      Make a note of responsiveness, meaning: How quickly does the interface respond when you attempt to use the interface components?

      Also, if the responsiveness is really bad, so terrible you can barely do anything and can't really measure or assess responsiveness, try reducing the number of graphs in the display.

        Attachments

          Issue Links

            Activity

            Hide
            pkulzer Paige Kulzer added a comment -

            IGB running in JDK21 and JDK8 seems to have the same level of responsiveness when working with large amounts of data. Once the data is loaded, both versions of IGB were similarly slow to select the graph tab, highlight all tracks, and apply changes to all tracks. Zooming in and out also takes a while at first, but then after zooming once or twice, IGB seems to speed back up again in both versions. Reducing the number of graphs present in each version seemed to help responsiveness in similar ways, but IGB still seemed useable (albeit a bit slow) with 20 graphs present in both versions.

            Show
            pkulzer Paige Kulzer added a comment - IGB running in JDK21 and JDK8 seems to have the same level of responsiveness when working with large amounts of data. Once the data is loaded, both versions of IGB were similarly slow to select the graph tab, highlight all tracks, and apply changes to all tracks. Zooming in and out also takes a while at first, but then after zooming once or twice, IGB seems to speed back up again in both versions. Reducing the number of graphs present in each version seemed to help responsiveness in similar ways, but IGB still seemed useable (albeit a bit slow) with 20 graphs present in both versions.
            Hide
            ann.loraine Ann Loraine added a comment -

            A question:

            Is it possible to identify a smaller number of loaded graphs that would let us observe a possible difference in responsiveness?

            Maybe a better strategy would be to progressively load one graph, two graphs, three, into IGB 9 and IGB 10. The point at which a noticeable lag appears might be different depending on the IGB version.

            However, it might be impossible to identify a difference.

            Show
            ann.loraine Ann Loraine added a comment - A question: Is it possible to identify a smaller number of loaded graphs that would let us observe a possible difference in responsiveness? Maybe a better strategy would be to progressively load one graph, two graphs, three, into IGB 9 and IGB 10. The point at which a noticeable lag appears might be different depending on the IGB version. However, it might be impossible to identify a difference.
            Hide
            pkulzer Paige Kulzer added a comment -

            In both IGB 9 and IGB 10, I added scale graphs one at a time and tested out the functionality of various tabs/tools in IGB after each addition. That looked like this:

            • Add a Scaled Graph and Load Data, noting any slowdowns that might occur here
            • Flip through all of the tabs at the bottom of IGB and see that they're opening/loading correctly
            • Click the Select All button in the Graph tab
            • Zoom in and out with the slider at the top of IGB
            • With the hand tool selected, drag the main view of IGB left and right
            • Zoom all the way out and repeat this process

            Both IGB 9 and IGB 10 start slowing down at about the same point (after four graphs are added and loaded). Both IGB versions also start freezing for a bit when adding a large number of graphs at once OR when working with more than 15 or so graphs at once.

            I didn't notice any notable discrepancies between the two versions in terms of responsiveness.

            Show
            pkulzer Paige Kulzer added a comment - In both IGB 9 and IGB 10, I added scale graphs one at a time and tested out the functionality of various tabs/tools in IGB after each addition. That looked like this: Add a Scaled Graph and Load Data, noting any slowdowns that might occur here Flip through all of the tabs at the bottom of IGB and see that they're opening/loading correctly Click the Select All button in the Graph tab Zoom in and out with the slider at the top of IGB With the hand tool selected, drag the main view of IGB left and right Zoom all the way out and repeat this process Both IGB 9 and IGB 10 start slowing down at about the same point (after four graphs are added and loaded). Both IGB versions also start freezing for a bit when adding a large number of graphs at once OR when working with more than 15 or so graphs at once. I didn't notice any notable discrepancies between the two versions in terms of responsiveness.
            Hide
            ann.loraine Ann Loraine added a comment -

            Thanks for doing this! I think we can conclude that the two versions of IGB are about equal when it comes to the user experience of responsiveness and speed.

            Moving to done.

            Show
            ann.loraine Ann Loraine added a comment - Thanks for doing this! I think we can conclude that the two versions of IGB are about equal when it comes to the user experience of responsiveness and speed. Moving to done.

              People

              • Assignee:
                pkulzer Paige Kulzer
                Reporter:
                ann.loraine Ann Loraine
              • Votes:
                0 Vote for this issue
                Watchers:
                2 Start watching this issue

                Dates

                • Created:
                  Updated:
                  Resolved: