Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Hello,
How would you recommend setting up an intermediary data model in Qlik Sense Enterprise?
It looks like binary loading isn't possible to make qvf to qvf connection. Any ideas for how to implement this, without causing memory issues? I would like to manage a data model separate from the dashboard, that would allow for several front-end developers
Thanks!
Binary loads are possible:
https://help.qlik.com/en-US/sense/August2022/Subsystems/Hub/Content/Sense_Hub/Scripting/ScriptRegula...
However, if you are unable to do the above for some reason, there is something else I would suggest. You could create QVDs for each table in your data model. Then you could create a QVS file that has the script to load those files in with the proper keys and links. The QVS file can be shared between all of your dashboards (your developers would just call the QVS file itself). Then you're only managing the script in one place instead of in each individual report.
Yes, you could have a different QVS file that chooses different fields for each QVF. Keep in mind that you will then need to maintain each QVS separately, so it will be more maintenance work in the long run.
Binary loads are possible:
https://help.qlik.com/en-US/sense/August2022/Subsystems/Hub/Content/Sense_Hub/Scripting/ScriptRegula...
However, if you are unable to do the above for some reason, there is something else I would suggest. You could create QVDs for each table in your data model. Then you could create a QVS file that has the script to load those files in with the proper keys and links. The QVS file can be shared between all of your dashboards (your developers would just call the QVS file itself). Then you're only managing the script in one place instead of in each individual report.
Thanks! I am waiting to hear if the binary functionality is allowed (US government server). Your work-around is interesting. I worry that the binary loading will exponentially increase the data storage requirements with a large data model. With your work-around, we could control the tables that are pulled in to developer QVFs, I believe, to keep from having all the data pulled into each QVF. Does that sound right to you? Thanks again.
Yes, you could have a different QVS file that chooses different fields for each QVF. Keep in mind that you will then need to maintain each QVS separately, so it will be more maintenance work in the long run.