posted

1 Comment

Usable capacity inside the VMware Cloud on AWS service depends upon the VM storage policy configuration.   This flexibility, while undoubtedly powerful, presents a challenge when answering the otherwise straight forward question of usable capacity. Traditionally the overhead associated with a particular parity scheme or data protection methodology is configured upfront on a container or volume or storage.   Since vSAN empowers the operations team to configure those attributes on a per-object basis, usable capacity depends upon said configuration.

Usable capacity

An additional consideration is the number of nodes in the cluster.  An entry scale three host cluster is restricted to the 1 failure – Raid-1 (Mirroring) Host Failure Toleration policy.  By the time we get to six hosts in a cluster, we have sufficient fault domains to support any policy configuration.  However, we also have to begin to consider host reliability.  Per the VMware Cloud on AWS SLA any Cluster with six or more i3 hosts in a single Availability Zone must use double failure protection to qualify for SLA credits.

Putting it all together, If we assume uniform policy configuration, we can project usable capacity in the service. Usable capacity

Stretched Clusters are subject to the same rules and restrictions as well as the additional requirement to maintain a local copy; a Stretched Cluster protects from AZ failure by keeping a copy of the data in each AZ.  While there is more to it, at a high level, a Stretched Cluster will provide half the storage of a standard cluster of equivalent size.

Usable capacity

The reconfigurability and variation of VM Storage Policy Based Management enable customers to shape their SDDCs consumption to match the requirements of their use case or workload. To simplify sizing and planning, VMware maintains a publicly accessible sizer and TCO calculator.

Availability

To view the latest status of features for VMware Cloud on AWS, visit https://cloud.vmware.com/vmc-aws/roadmap.

Resources: