Hello, friends! I am currently working with a friend in a project involving a QlikView dashboard.
In this dashboard (with 15 different reports) we have noticed that the loading time of every report varies depending on the previously accessed reports.
Let's say I restart my services to empty my cache, and I measure how long does each report take to load, accessing them in the order A > B > C > D.
Then I restart my services again and make the same measurements but accessing the reports in the order D > C > B > A.
My results show that a 14 seconds load time can become 12 seconds (taking the average measure from 3 tests to avoid the obvious spread), which leads me to think that there is some kind of "shared resource" which is loaded in the first reports accessed.
How can I know exactly what are the reasons for this "reduced loading time"?
I already knew about the cache functionality, but my question is, having these reports that apparently do not share any calculations, how can I check what elements are shared? Is there anything more than calculations (not shared between these reports) that could be shared and thus reducing load times for the latest reports accessed?
I don't think that you could see anywhere which results are cached unless you goes through all your objects and checks the used dimensions and expression. But you could measure how many RAM the various tables/fields and objects consume and how long the calculations need - for this take a look here: Recipe-for-a-Memory-Statistics-analysis.