Qlik Community

Qlik Sense Deployment & Management

Discussion board where members learn more about Qlik Sense Installation, Deployment and Management.

Announcements
BI & Data Trends 2021. Discover the top 10 trends emerging in today. Join us on Dec. 8th REGISTER
cancel
Showing results for 
Search instead for 
Did you mean: 
Highlighted
Creator III
Creator III

Why opening time of an app or a sheet is always way much faster on an unpublished app ?

Hi,

 

I have a very specific question 🙂 

 

  • We have many apps, a few big apps (>1Gb), a very few very large apps (>15Gb).
  • Some apps/sheets are long to open, for the first time that's normal --> it's going in cache.

 

But we found that opening these apps from the "My Work" area was always VERY much faster to open (first time and also the times after).

 

  • I am currently testing to disable all "custom" security rules and replacing them with only one very specific rule like : if user = Me then grant access to all actions on all resources.
  • And still, this makes no diffference when opening published apps.

 

 

Does somebody know why is the "owner" (read-only rule) is so much faster to render ? Is it hard coded or something that makes Qlik to bypass the evaluation ?

Or why in general is it much faster to open an app from "My Work" ?

 

Thanks for any input !

Regards,

Thomas

Labels (4)
2 Replies
Highlighted
Employee
Employee

This is a fairly broad topic, to be honest. What version of Qlik Sense Enterprise are you testing on? I ask because there is a changeset in June 2020 which is very relevant here:

Improved performance for resolving linked master objects in OpenApp
Jira issue ID: QB-1426
Description: When resolving linked master objects in sheet objects, OpenApp exhibited poor and undue
performance scaling with the number of sheets and sheet objects present in the app. This has been resolved
and the proper object indices are now used in this lookup, taking practically negligible time

The crux of this issue was inefficiencies when opening an app with a lot of sheets with master items used in those sheets. This would match the pattern that you have here.

The other one you've already isolated as not at play is very broad security rules can cause slowness (i.e. a user who can see all user's personal sheets on published apps).

Highlighted
Creator III
Creator III

Hi Levi,

Thank you for your help, and sorry for the delay I wasn't at the office.

  • We are currently on QS June 2020 Patch 2.

That's still a good idea to test removing master objects/dims/measures or other content as we also have quite many variables / bookmarks in these apps. Maybe we could eliminate a few latency potential causes from our test.


Regards,

Thomas