vCloud Director allocation models and high availability

[joeflintxxxxx]

Hi, i've created a provider vDC and Organisation

Next to create an ORG vDC using the Flex Model.

By default the total pvDC compute is displayed (CPU & memory) i.e. matches that inthe system created resource pool.

I'm assuming therefore, vCD does not account for ESXi host failures and if we want to allow for at least one host failure.

Hence if a host has 32CPU (or 150GHz) and 512GB memory, then then we need to reduce this capacity from the overall total assigned to the vDC. Plus an additional amount if we only want 80% maximum utilisation.

Does this make sense, does anyone have any experience is assigning compute that they can share?

thanks

[Sreec]

Yes, you are right about Host failure scoping. We need to ensure vSphere HA is configured accordingly. Flex model gives you the feasibility to allocate resources for tenants and reserve any compute per VM level if required. So you need to consider all the use cases and business requirements before sizing the platform.