Re: [Openstack] How to limit the total virtual processors/memory for one compute node?

2011-05-24 Thread Joseph Suh
Huang, If you are willing to modify code, you might want to take a look at the code at lp:~usc-isi/nova/hpc-trunk that has a scheduler (nova/scheduler/arch.py) that does not allow creating new instances if cpu or other resources are used up. If you have any question on the branch, please feel

[Openstack] How to limit the total virtual processors/memory for one compute node?

2011-05-23 Thread Huang Zhiteng
Hi all, In my setup of Cactus, I found Nova scheduler would place newly created instance to a compute node that is already full occupied (in terms of memory or # of virtual processors), which lead to swapping and VP overcommitting. That would cause serious performance issue on a busy

Re: [Openstack] How to limit the total virtual processors/memory for one compute node?

2011-05-23 Thread Lorin Hochstein
Hi Huang: You can use the simple scheduler, allocates new instances to hosts that have the fewest instances currently running. --scheduler_driver=nova.scheduler.simple.SimpleScheduler More sophisticated schedulers are currently under active development, but they haven't made it to the trunk