Separate cluster or resource pools? How would you do it?

[ejward]

Here's my situation. I've just set up 2 ESX hosts in a cluster with shared storage. lets call this the production cluster. The same department also bought a 3rd ESX host as a sandbox server. All of these servers already see the same SAN storage. Should I keep the sandbox server seperate or, just put it in with the other hosts and just have a sandbox resource pool?

[sbeaver]

I would combine and create a resource group for the sandbox. You get more resources and failover capacity and the sandbox gets the resources it needs. Win Win

Steve Beaver

VMware Communities User Moderator

*Virtualization is a journey, not a project.*

[aguacero]

I agree with Steve. Add it to the existing cluster and create a "Sandbox" Resource pool. It provides another additional server for HA the production environment.

If you found this information useful, please consider awarding points for "Correct" or "Helpful". Thanks!!!

[azn2kew]

It depends how your management wants to do between production/dev/test/staging environments. I've allocated the clusters dedicated for production/dev/test servers themselves because the environment is fairly large and we have the hardware host and maintain its own role and functions. But if you have 3 ESX hosts than combining it using resource pool to maximize your resources allocation. I still wouldn't recommend combining neither production/sandbox into my DMZ cluster though that's pretty strict and protected.

If you found this information useful, please consider awarding points for "Correct" or "Helpful". Thanks!!!

Regards,

Stefan Nguyen

iGeek Systems LLC.

VMware, Citrix, Microsoft Consultant

[ejward]

Oh sure, bring up security. Well .... lets just say that this is in no way going to make things worse security wise.

[RobBuxton]

Another item that can influence the decision is licensing. Some vendors insist that their product be licensed by physical CPU. That doesn't fit into the VMware cluster very well and having a separate cluster can get around that.