Preparing target site

Where you place Datamotive nodes on the recovery site determines what you can recover, and how many workloads you can recover in parallel. Prepare the target site against the rules below before creating sites and protection plans.

Node placement rules

• The Management Server and Replication Nodes support a replication source or target within one cloud region (or one vCenter Server for VMware).
• When the source is a cloud platform, deploy the Datamotive nodes in the same region as the instances being protected.
• When the target is a cloud platform, deploy the nodes in the region where the protected instances are expected to be recovered.

Multi-AZ recovery

If recoveries will span multiple availability zones within the target region:

• Deploy the Management Node in any one availability zone. On Azure, deploy with the No Zone resiliency option — this enables recovery into multiple zones.
• Deploy Windows Prep / Replication Nodes in each availability zone where recoveries are expected. Size the count by parallel recoveries: each node supports recovering 20 VMs in parallel in its zone.

VMware as recovery site

Two additional placement requirements apply:

• The recovery-site Datamotive nodes must reside in the same datacenter used in the recovery configuration of the protected VMs.
• The nodes must have access to the disks of the recovery VMs — the datastores where VMs will be recovered must be accessible to the Datamotive nodes.

AWS Dedicated Hosts as target

Deploy Datamotive nodes in the same region as the target Dedicated Hosts. The nodes themselves can run on shared-host EC2 instances — there is no need to place them on the Dedicated Hosts. If the Dedicated Hosts span multiple AZs, deploy a Prep/Replication node in each AZ to facilitate recovery there. Note the Dedicated Host limitations.

Node sizing on the target

Instance sizes, storage types, and image names per platform (VMware, AWS, Azure, GCP) are listed in the Nodes page and the sizing guidelines. As a rule of thumb: Management Server, Replication, and Prep Nodes run 4 vCPU / 16 GB RAM with 50 GB storage; DeDupe Nodes need 500 GB (SSD preferred) sized by anticipated unique data.

Sandboxed environment for test drills

Test drills recover workloads from the last consistent replicated copy without disrupting production or replication. To keep test-recovered VMs fully isolated:

See DR drill for the drill workflow and cadence, and Recovery testing for the test-recovery procedure itself.

Related docs

• Nodes

• Sites

• DR drill

• Unsupported configurations