How can I prevent one container from consuming all of the CPU capacity, preventing other workloads from running properly?
Use the resource management features of Solaris. This requires using some combination of the Fair Share Scheduler, CPU caps, assigned (’dedicated’) CPUs, and/or [Dynamic] Resource Pools features. Fair Share Scheduler: FSS guarantees a minimum amount of CPU utilization, so it doesn’t waste CPU cycles. Excessive CPU use is only prevented if there is contention for CPU resources. Minima are specified by “shares” and enforced by the Fair Share Scheduler. You can assign 100 shares to each of two workloads, 200 shares to a third workload, and 400 shares to the global zone. The first two workloads will get at least 1/8 of the CPUs’ capacity, if they need it, and the global zone will get at least 1/2 of the CPUs’ capacity, if it needs it. If only one Container wants to use the CPUs, it will be able to use all of the processing capacity of the system. You should give the global zone a quantity of shares similar to the largest quantity given to any Container, to ensure that you can manage the Co
Related Questions
- I have Printopia installed on Mac OS X Server, and its not working properly. Anything I should know about running Printopia on Mac OS X 10.6 Server?
- Why is it that System Monitor and other CPU monitors indicate that my CPU is 100% busy when running CPU Cool?
- How do I know that the SingTel Online Backup program is properly installed and the service is running?
