Comment 5 for bug 1835958

>There are two things that _should_ help:
>
>- Disable power state checks in nova.conf.
>- Break up your ESXi clusters into smaller ones.

With as much churn our environment has, disabling the power state checks wouldn't be good. We use it for dev/test, and the users do a lot of stuff within the instance operating system. We are changing the periodic check to 1 hour rather than the default 10 minutes, though.

Breaking up the large clusters into smaller ones will likely help a ton but will really mask the underlying issue: actions taken on a specific instance causes a _full_ power sync to be kicked off for every single instance.

I know this would mean a pretty involved rewrite, but an action taken on a particular instance should only do a power sync for that particular instance rather than every single instance. Also, I'm not understanding the need/benefit of doing a whole power sync when creating instances. Scaling up a cluster would perform better if that isn't necessary.

Thanks!
Tom