Azure Site Recovery

Workload support

Provider-documented

Supported scenarios include Azure VMs, on-premises VMware VMs and physical Windows/Linux servers, plus Hyper-V VMs. Each has a separate support matrix; the VMware/physical path requires supported 64-bit guests and compatible disk and machine configurations. [1] [2] [3]

Recovery plans group machines, define startup order, and allow automation to coordinate application recovery. [6]

Recovery locations

Provider-documented

Azure-to-Azure recovery uses a selected target Azure region. Microsoft describes global cross-region recovery but applies restricted-region rules; for example, Switzerland West is reserved for Switzerland North customers. [4]

For Hyper-V, Microsoft states that customer recovery data remains in the configured target region; a Recovery Services vault in another region holds metadata rather than that customer data. Hyper-V has its own target-region restrictions. [3]

Testing scope

Provider-documented

A test failover can recover one VM or a multi-VM recovery plan without interrupting ongoing replication or production. You choose a recovery point and test virtual network, inspect the recovered environment, record observations, and clean up test VMs afterward. [5]

Cost details

Provider-documented

Microsoft bills by protected instance, with no Site Recovery service charge for that instance’s first 31 days. Storage and transfer charges can still apply during that period. The monthly rate was not rendered in the retrieved pricing table, so no current dollar figure is asserted here. [7]

Budget beyond the service fee: replicated disks, recovery-point snapshots, storage operations, applicable outbound transfer, and VM compute during drills or recovery. Software in the recovered machines also needs appropriate licensing. [7]