Q: Do we have some hardware set up recommendation for Qlik Sense? for example QlikView is not friendly with NUMA enabled etc.
A: Yes. Since Qlik Sense Engine share a lot of common characters with the QlikView Engine, therefore the server settings recommendations for the best performance can be applied to both products. You can find those recommendations from the community post at
Q: Does the data 3rd type of data loaded in RAM (application cache) is shared between users ?
A: Yes, the cache memory is shared among users, but not among nodes
Q: If a user has access to only half an app's data, when they open that app, does Qlik Sense load all data to memory? Basically, what's the impact on RAM?
A: When a user opens up an app having Section Access, the complete data set for the data model will be loaded into the RAM. However, the user will only see the subset of the data that is allowed for him/her.
When another user accesses the same app, no further data load for the App data model is needed since all data set is already loaded into RAM.
Q:In a multiple node configuration, will the applications utilize core across the nodes
A: Qlik Sense does not share nodes for an individual App instance. If an App will be accessed from multiple nodes, multiple instances will be loaded separately on each node. User sessions will be load balanced according to the rules.
Q: And can QS share one big app on many QS engine of cluster?
A: No. Qlik Sense does not share CPU or RAM resources among nodes for an individual App. Apps loaded on a particular Engine node will only consume the RAM on that node.
Q: How much worse is it to have a virtualized server?
A: Qlik Sense fully support the deployment on virtualized environment. However, special care should be taken when considering the virtualization for any memory and CPU intensive Apps and tasks.
I can’t find any updated white paper for Qlik Sense about the performance in virtual environments. But this one for QlikView Server has some very useful information regarding to the Server Virtualization and Qlik Engine performance on VM.