topic Re: Application Run Time (i.e. date stamp Methodology) in App Development

Application Run Time (i.e. date stamp Methodology)

crichter14 — Wed, 08 Sep 2021 13:15:57 GMT

Ok this is more a best practice question than a Qlik question. I feel like if I knew the exact terminology I might be able to research it better.

So our Qlik application gets it's data from multiple sources. They update at different times. So let's say a 'simple' SALES table has over 240 sources. Most are smaller but that gives you a general idea of what how big the data set can get. That data is consumed in SSMS and sent over the Qlik Sense in a optimized format.

The problem we're having is when we put a SINGLE date in the Qlik Sense dashboard as the 'updated' time we're in disagreements on what time to use.

Some sources update 1x a month, that makes the data look out of date and the users panic. Some sources may have failed in their update so for those we go back to the prior successful run. The user wants to know when the source was pulled not just when the Qlik Sense app reloaded (which would be easier to do).

Here's the question, methodology wise, is there a best practice on what source timestamp to display in the dashboard as the 'modified' date when you have multiple sources?

Re: Application Run Time (i.e. date stamp Methodology)

NadiaB — Mon, 13 Sep 2021 19:20:32 GMT

Hi @crichter14

It looks to me that you probably need to make a decision about what to display, I would probably add an extra field that shows how often the data is reloaded, so when the customer sees a date (for the example that you mentioned that the data is reloaded each month) the user can be aware about how often is expected to see that information updated.

Hope it helps.

Re: Application Run Time (i.e. date stamp Methodology)

stevejoyce — Mon, 13 Sep 2021 19:29:55 GMT

I think this I did something similar to what you are asking about...

Agree, last reload date doesn't indicate when source data was pulled if you have multiple layers/parts of reload tasks.

In each of my extract qvf's, i add a variable at the end let vTime = Now(); And store into a table with document name/scope of data name + $(vTime). Your front-end app can load the latest records for each "scope of data name". If you have too many sources/times to display simply in your app, you can have an "About" sheet at the end of your app that contains this info. A table of your source times for each source system/reload.