Performance Issues with Large KML Files in Qlik Sense (On-Prem & SaaS)
Hi everyone,
I’m currently working with large-scale geospatial data in Qlik Sense, both on-premises and in the SaaS version. My dataset includes global country, state, district, and municipality/ZIP code area layers in KML format, which are highly detailed. While I can visualize these layers on the map, I am experiencing significant performance issues.
Surprisingly, the performance is almost identical in both Qlik Sense SaaS and on-prem, despite running on a very powerful server (which only utilizes around 20% of the available RAM). The responsiveness of the maps is quite slow, making the user experience less than ideal.
I would really appreciate insights on the technical limitations of Qlik’s mapping engine:
What are the recommended or maximum file sizes for KML layers?
Are there best practices for optimizing large geospatial datasets in Qlik Sense?
Are there specific configurations or optimizations that could improve performance?
If anyone has dealt with similar challenges or found ways to efficiently handle detailed KML data in Qlik Sense, I would love to hear your suggestions!
Just a thought - the issue may not directly be caused from the backend else the frontend in regard to the amount of the transferred data and/or to render them. It may like be displaying a larger table in which the relevant sub-set of data is often much faster calculated as transferred and/or the html column/row structure is rendered.
If this is really your biggest bottleneck you may consider to use n versions of your KML with a different granularity and depending on the context and/or time the most suitable one is used - like many web-sites providing pictures at first with a low resolution and after a while they are replaced with a higher one.
Before going in this direction makes sure that the data-model and the UI expressions are as much as possible optimized.
