Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Hi,
We have two nodes that run the data loading jobs. We have seen that one node is having much more load than the other one.. CPU usage is pretty much identical but ram usage is not. One node ram usage reaches the maximum while the other ram usage is not even half.
What is the way of doing a manual load balancing so that we can setup half of the big apps to be reloaded on one node and half on other
Hi,
To dedicate applications to be reloaded on specific node(s), you will need to change the default load balancing rule in the QMC.
Here is an example using custom properties: https://adminplaybook.qlik-poc.com/docs/asset_management/apps/review_pinning_load_balancing.html
When running task in chain, the entire chain will run on the same Scheduler node.
Hope this helps!
Kind regards,
Bastien Laugiero
Hello, are both nodes identical in terms of performance (CPU and Memory)? Are your tasks running in chain?
Hi,
To dedicate applications to be reloaded on specific node(s), you will need to change the default load balancing rule in the QMC.
Here is an example using custom properties: https://adminplaybook.qlik-poc.com/docs/asset_management/apps/review_pinning_load_balancing.html
When running task in chain, the entire chain will run on the same Scheduler node.
Hope this helps!
Kind regards,
Bastien Laugiero
hope you're well !
Are you sure about this sentence : "When running task in chain, the entire chain will run on the same Scheduler node. " ?
Or maybe it is not the case anymore on newest Qlik Sense versions, i have an exemple (August 2021) in a multinode architecture, where 2 differents apps are reloaded on 2 differents nodes even in a task chain.
What i'm looking for, is a way of "fixing" a specific node when excuting an external task (which we can't pilot with loadbalancing rule). Woulld you have an idea ?
Thanks a lot and have a good day !
Johann