Skip to main content
Announcements
Have questions about Qlik Connect? Join us live on April 10th, at 11 AM ET: SIGN UP NOW
cancel
Showing results for 
Search instead for 
Did you mean: 
allon_s18
Contributor
Contributor

How to use QV as a in-memory data source ?

Hello!

In my company we already use QV as a in-memory report tool which keeps stocks. Stocks are in in-memory.

I have another IT system which should get data from QV very quickly.

My questions is:

1) is it possible ?

2) if yes then how to implement it - can I read data directly from in-memory or from QV file ?

Picture below shows two Options:

Option-1 - read data quickly from in-memory

Option-2 - read data slowly (???) from QV file.

Please help.

BR, Alex.

 

 

 

 

6 Replies
tresesco
MVP
MVP

...another IT system ...

Is it  also qlik environment? If not, I guess it would not be possible to utilize 'in-memory'/ read-from-qv file. 

marcus_sommer

Qlik has a very specific data-structure so that your IT system would need to be capable to address it in the same way like the Qlik engine does to access the in-memory data. Therefore I doubt that there is a real way especially not only by considering the technically aspects else also the legal ones.

Therefore I think you will need at least one layer between Qlik and your IT system. For this you could use any file-exchange or any write-back approach to a database or maybe a connector like this one: tiq-solutions.de - tiq-java-service-connector. You could contact the head - Ralf Becher - behind the connector to find out if it might be possible / sensible in your usecase.

- Marcus

rbecher
MVP
MVP

Hi Alex,

I think there are perhaps some products for getting data out of QV in-memory models but I have my doubts that works well.

What I would do instead is to run a Qlik Sense server beside where you binary load your QVW, or having an app with same load script.

Then you can use the engine API to access the in-memory model by defining a HyperCube. There is also an open source project q-risotto I started which provides REST API wrapper for this to make it easy:

https://github.com/ralfbecher/q-risotto

- Ralf

Astrato.io Head of R&D
allon_s18
Contributor
Contributor
Author

Ralf, colleagues.

I appreciate your quickest response.

Thank you very much for your valuable help.

Sincerely yours, Alex.

allon_s18
Contributor
Contributor
Author

Hello!

Another question.

QV has a report which I want to use as data source.

To export data to CSV I want to use NPrinting job from QV.

Is it possible to do it for QV-report which consists of 200 000 rows and 1200 columns ?

Are there some limitations and constraints for QV NPrinting ?

BR, Alex.

marcus_sommer

I could imagine that you run into various problems with such a big table with or without NPrinting in regard to limitations/bugs within the object, the RAM consumption and the calculation/transfer-times. I wouldn't do a try in this direction else I would create this table within the script and store it as csv.

- Marcus