Skip to main content
Announcements
Qlik Connect 2024! Seize endless possibilities! LEARN MORE
cancel
Showing results for 
Search instead for 
Did you mean: 
David_HealthDS
Contributor
Contributor

Qliksense and R integration between virtual machines

Hi all,

None of this is implemented yet so trying to use the forum to get a sense of considerations. Currently I have Qliksense running on a virtual machine in Azure. Separately I am going to set up another new virtual machine in Azure dedicated to data science work. Ideally this VM will only be used sparingly (i.e. not running 24/7) to reduce costs/maximise processing power.

I would like to set up a SSE extension between these virtual machines to integrate Qliksense and R, so that some of the advanced analytics work can be incorporated into Qliksense (e.g. forecasts, model predictions).

Can anyone provide any background on what requirements this will place on the new data science VM? I.e. if the Qliksense machine is operating 24/7 for end users does this necessitate the data science virtual machine to also be operating 24/7? Or are there ways around this in the way Qlik evaluates R scripts?

General advice welcome.

Thanks,

David

1 Solution

Accepted Solutions
Damien_Villaret
Support
Support

Hello @DA 

The configuration for analytic connection in the QMC is pointing to where the SSERPlugin is installed.

It's in the SSERPlugin configuration file that you will specify where the R server is.

https://github.com/qlik-oss/sse-r-plugin

If the issue is solved please mark the answer with Accept as Solution.

View solution in original post

3 Replies
Damien_Villaret
Support
Support

Hello @David_HealthDS 

Basically, if you are evaluate R expression in the Qlik Sense script and store it as data in the Qlik Sense app, then the R server won't need to be running when users are accessing the Qlik Sense apps (only during the app reload)

However, if you are evaluating R expressions in the Qlik Sense front end, then the R server would need to be running when users use the app.

If the issue is solved please mark the answer with Accept as Solution.
David_HealthDS
Contributor
Contributor
Author

Hi @Damien_Villaret 

That makes perfect sense i.e. if the required R calculation is static then the VM only needs to be on when app is being loaded, vs a dynamic calculation through user input would require a constant connection.

Happy to consider this closed but a slightly related question. If the R server is on a separate VM to the VM that Qlik is set up on, are there any dependancies on the R virtual machine with respect to Qlik? I'm just thinking of how analytic connections are set up regarding host and port configurations, as I don't think it will be the 'localhost' style scenario that is in most of the help guides. Example screenshot below.

Thanks,

 

David

Damien_Villaret
Support
Support

Hello @DA 

The configuration for analytic connection in the QMC is pointing to where the SSERPlugin is installed.

It's in the SSERPlugin configuration file that you will specify where the R server is.

https://github.com/qlik-oss/sse-r-plugin

If the issue is solved please mark the answer with Accept as Solution.