2.1 Disaster Recovery

Customer with on-premises or co-located vSphere environment is looking for a simple DR solution that would allow the replication of virtual workloads to a multitenant public cloud service. In the case of a complete loss of the vSphere infrastructure, the tenant can power on all replicated workloads in the public cloud with minimal loss of data (RPO = 15 mins).

1. Customer purchases cloud Organization Virtual Data Center (Org VDC) with DR option enabled. This can be augmented with additional Org VDCs that are not enabled for DR to provide capacity to run permanent workloads that are protected by solutions other than storage-level replication (for example, Active Directory servers or databases).

2. Customer sets up cloud networking infrastructure with Organization VDC networks that use the same subnets as the on-premises or co-located environment. Optionally, additional networks can be created for test failover purposes (isolated networks).

3. Customer configures a new vCloud replication endpoint in the VMware vSphere Web Client with its vCloud Director credentials.

4. Customer maps the vSphere networks in vSphere Web Client with vCloud Org VDC networks for failover and fail back purposes.

5. Customer enables protection in vSphere Web Client for each workload that must be protected. Recovery Point Objective (RPO) and number of point-in-time snapshots (vSphere 6 feature only) must be specified. RPO can vary from workload to workload. Guest quiescing and replicated data compression are supported as well.

6. Initial sync for each protected workload must be done before the workload becomes protected. This could be sped up with pre-seeding of cold clones (for example, through vCloud Connector with Offline Data Transfer).

7. Customer monitors vSphere Web Client replication dashboard to see if all protected workloads satisfy the required RPO and are in compliance.

• Customer can at any time initiate in vSphere Web Client test failover for given subset of protected workloads. These will be connected to vCloud Org VDC networks designated as test networks and powered on.

• In case of loss of on premises or co-located vSphere environment, the customer can initiate failover. Replicated workloads will be connected to the vCloud Org VDC networks designated as recovery networks. Failover can be initiated from within vSphere Web Client or alternatively via vCloud API.

• After the rebuild of the customer on-premises vSphere environment, the customer can reverse replication to prepare the workloads for a failback (only supported with vSphere 6.x on-premises).

• Metering for VMware licensing is based on number of protected virtual machines.

• Customer billing is based on resource usage (CPU, memory, disk space).