I've read a lot of articles on this and have seen this done many different ways. Please clarify and comment on the best method to accomplish this.
We have 4 ESXi 5.0 hosts in 1 cluster. I want to implement resouce pools so we can divide our CPU and memory resouces between production vms and test/dev vms. I'd like to keep it at about 80% for production and 20% for test/dev.
1. Create one resouce pool under the cluser called Dev, assign it low shares, then put the dev machines in that. Then all the production machines would reside at top level in cluster and get full resouces?
2. Create two resource pools. One for Dev with low shares and one for Prod with high shares. Move all machines into one or the other.
3. Create one resouce pool under the cluster named Prod, then a child resouce pool under the Prod resouce pool called Dev?
Any better methods?