CPU is reaching 90%(qlik sense engine is taking almost 80-85% cpu)
HI experts,
We are facing wired issue from past 2-3 days that actually we are having 1 central node and 1 rim node.and we have exactly 24 qlik applications of modern size 800mb to 2gb maximum in qmc.everytie central node is showing just 3% cpu & 5% memory in task manager but whereas for rim node it is showing 90% cpu and 2% memory.We observed apps performance is very very dull and not opening at all. what are the steps i need to take care while reloading the apps and how to reduce cpu in RIM node
It sounds like your Qlik Sense environment is experiencing an imbalance in workload distribution between your Central Node and Rim Node. The fact that the Rim Node is reaching 90% CPU usage while the Central Node remains underutilized suggests that most of the resource-intensive tasks, such as app reloads or user sessions, are being directed to the Rim Node.
To address this, the first step would be to review the load balancing rules in the Qlik Management Console (QMC). These rules determine how app sessions and reloads are distributed across nodes. It's important to ensure that the Central Node is also contributing to the workload rather than leaving everything to the Rim Node. Adjusting these rules to better balance the load should help alleviate the pressure on the Rim Node.
Another critical area to investigate is the task scheduling. If multiple reload tasks are running simultaneously on the Rim Node, this could easily overwhelm its CPU. Staggering the reload schedules or redistributing these tasks across nodes can prevent resource contention and improve overall performance.
Given that your apps are quite large (800 MB to 2 GB), optimizing them could also make a significant difference. Removing unnecessary fields, tables, or unoptimized scripts from the apps can reduce their size and computational demands. Testing optimized versions of these apps in a staging environment before deploying them can help ensure they perform well.
Finally, it’s worth checking the hardware specifications of the Rim Node. If the CPU is consistently maxing out while memory usage is low, it might be time to upgrade the node’s CPU capacity or add more Rim Nodes to distribute the load further.
By making these adjustments, you should see an improvement in the performance of your Qlik Sense environment. Let me know if you’d like more specific guidance on implementing these changes!
