Performance tips for large lists

Thread is closed for posting
3 posts, 0 answers
  1. Erik
    Erik avatar
    24 posts
    Member since:
    May 2011

    Posted 25 May 2012 Link to this post

    I have a jump panel right now that has anywhere from 1000-6000 items. Each item have about 4-5 controls. (Stack panels, images, text). I've found that saving state with this many controls is noticeably delayed. 1-5 seconds. I have tested this on a Samsung Focus and a Lumia 900. You don't really see it on the emulator because, we'll it's pretty darn fast. But on the phone moving from page to page, or resuming from tombstone is slow at times.

    Is there a way to make state/tombstoning just ignore the requests so it doesn't try to remember where all 10K controls are? This obviously breaks some standards and can break the UI. I am still curious on thoughts about this.
  2. Deyan
    Deyan avatar
    2147 posts

    Posted 28 May 2012 Link to this post

    Hi Erik,

    Thanks for writing and for your questions.

    Based on the details you are sharing I am not quite sure that I correctly understand the case. Normally, you do not have to store information about where your controls are etc. In the general scenario when an application is thombstoned you should store application state information like user input, the current page of the app at the time of thombstoning, the scrolling position of listboxes and so on.

    Can you please share some further details so that we can make sure that we correctly understand the case and thus be able to further assist you?

    Thanks for your time.

    the Telerik team

    Explore the entire Telerik portfolio by downloading the Ultimate Collection trial package. Get it now >>

  3. Erik
    Erik avatar
    24 posts
    Member since:
    May 2011

    Posted 05 Jun 2012 Link to this post

    We can close this thread, I think the issue is there is a first chance exception being thrown on session state load which slowed down my emulator because the list had over 5K entries. 

    After some code clean up, it is now relatively speedy.
Back to Top