So right now we have a jenkins configured with over 5GB HEAP and still it takes about 20 minutes to render the dashboard.

Afaik from my research the dashboard does not use any caching anymore. So this creates big issues on high load machines.
We have a SSD system with about 100.000 IOPS possible and the machine is taking ages.

Is it not better to use caching and to decide a user if he wants to wait for a forced (by user click) update of the cache? Or maybe to do it by a trigger in the times the jenkins is on low load e.g. in the night?
Is there a way to reenable the caching for the builds again?

Right now we cannot use jenkins and the dashboard anymore, since it is unusable slow. Hope someone can help us.
Here (attached file log) what we could find out via /threadDump URL of jenkins. Hope it helps