Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Hi Qlikers,
I'm working on Qlik Sense Desktop and some of my apps are very slow. The apps size is around 30Mo and once a day it will just freeze my computer. I have an other App that is about 300Mo and there is no problem.
Basically this I what is happening. If I have a filter selected on a small part of my data there is no problem. When I remove all my filters the update of my charts can take about a minute or freeze my computer. Also just so you know this app had a size of 180Mo before, so reducing it to 30Mo didn't help.
What are the good practices to make an App efficient? Is it due to my data base structuration, my functions, .... Is there already a post, document related to this topic?
Thanks
Xavier
1. Data source - Yes, you can apply where clause and filter the required data.
2. QVD - using STORE keyword, it is the compression technique, may be it is the same QVF
3. Section access - displaying only required data to users
4. Incremental is instead of full refresh just update only the new entries
Thanks,
SV
Few suggestions would be:
1. Data Source Filters - Directly reducing the data and getting the required ones
2. Remove unused fields
3. Create QVD files
4. If possible perform partitioning or section access to see only the required data by the users.
5. Performing incremental loads whenever possible
Double check the data model in both the files and try to see if you have got issues with the model itself
Thanks,
SV
to add on to what S V had to say, if you have objects that do not need to show data unless something is slected, you can make those expressions contional based on certain selections. That is a typical practice when showing details - you do not want to show every record when something is not selected, for example
Thanks for your quick reply.
So if I resume:
1. Data Source filters. This means that in the script I put some conditions that will only load the required data for the app I'm building?
2. Fields already reduced to the minimum I need
3. QVD file the same as QVF?
4. Partitioning or section access? I'm not really familiar with this...
5. Incremental loads, also not familiar with this
6.I will also check my data model.
Adam, I think that your suggestion is also something I have to work on. What is happening is that my bar charts are "quickly updated" but a table with data can take minutes... I will look if I find a post about how to do it.
Fo sure something is wrong with my new app, because there is not that much data. So I guest I build it wrong...
Thanks for your help Qlikers
1. Data source - Yes, you can apply where clause and filter the required data.
2. QVD - using STORE keyword, it is the compression technique, may be it is the same QVF
3. Section access - displaying only required data to users
4. Incremental is instead of full refresh just update only the new entries
Thanks,
SV