VMware Cloud Director service instances have built-in resiliency for availability of the service provider and tenant portals. You can configure VMware Cloud on AWS clusters to ensure high availability of tenant workloads in VMware Cloud Director service.

VMware Cloud Director service Deployment Model

Each VMware Cloud Director instance is deployed across multiple AWS Availability Zones in its region.

If one of the availability zones experiences an outage, new nodes are automatically deployed in another availability zone to replace the nodes that were impacted by the outage, ensuring that the VMware Cloud Director instance remains operational.

VMware Cloud on AWS SDDC Configuration Considerations

As a service provider, you can determine whether to use a standard SDDC configuration or a stretched cluster SDDC configuration for your tenants.

To provide an extra layer of resiliency and to guarantee the availability of the VMware Cloud Director service workloads, you can choose to associate your VMware Cloud Director instances with stretched cluster VMware Cloud on AWS SDDCs to protect against availability zone failure.

If you choose a stretched cluster SDDC configuration, a vSAN stretched cluster is deployed across three availability zones. This results in the creation of a vSphere cluster that can survive the loss of an entire availability zone. To protect against split-brain scenarios and to help assess site health, a managed vSAN witness node appliance is also created in a third availability zone. The third availability zone is picked at random from the remaining availability zones.

AWS Region Outage Impact to VMware Cloud Director Instances

In the unlikely event of a failure of an AWS region that hosts VMware Cloud Director instances, both provider and tenant portal access, as well as any configured access to VMware Cloud on AWS SDDC instances within the impacted region will become unavailable.

VMware Cloud on AWS SDDC instances associated with the impacted VMware Cloud Director instance but not in the affected region will remain operational and accessible to the provider and tenants outside of the VMware Cloud Director instance portals.

VMware Cloud Director instances in regions that are not impacted by the outage but are configured with multisite pairing to the impacted VMware Cloud Director instance will continue to operate with impacted resources remaining temporarily unavailable.

VMware Cloud Director Instance Disaster Recovery Plan

Regular backups of all VMware Cloud Director instance databases are stored in separate AWS regions. In the event of an extended region outage, and if the time to restore service is sufficiently faster than the anticipated time for outage recovery, VMware Cloud Director instances will be recovered in a new AWS region.