I have a node with 8-core 2GHz CPU. It runs 128 gears, and it turned out that this node runs out of CPU. So I replaced the CPU with a new one 8-core 3GHz. The problem is that CPU usage is limited as described in Admin Guide (https://docs.openshift.com/container-platform/3.3/admin_guide/quota.html). And in this guide CPU limits are measured in CPU core percentage. Thus after CPU is replaced every application instance gets more CPU time. But application instances don't need more CPU time because applications have not changed. I'd better share additional CPU time between new instances.
In real world some apps need to run prolonged calculations sometimes, like cron jobs. I need to limit CPU usage by those apps to 500MHz in order neighbor apps to have enough CPU time.
Is there a way to specify CPU limits in MHz?