Azure Site Recovery (ASR) is a disaster recovery solution for physical servers. The physical server–based environment can be replicated to Azure using ASR. The same approach can be used for migration, where the replicated environment is failed over and then continues to run from Azure. An additional VM or physical machine should be available on-premise to set up the ASR components.
Azure Backup does scheduled backup of VMs at specified time intervals to address the data and application resiliency concerns of environments hosted in Azure. If you have stringent recovery time objectives (RTOs) and recovery point objectives (RPOs) defined in a business continuity and disaster recovery (BCDR) strategy, Azure Site Recovery (ASR) could be the better choice because it replicates data continuously to a different region in Azure. Note that this service is in preview at the time of writing this book. It is recommended to include ASR in the architecture to enable environment-level resiliency once the service becomes generally available. ASR for Azure VMs can be enabled directly from the Azure portal with minimal configuration steps. While Azure Backup requires the vault be in the same region as IaaS VMs, ASR mandates that to enable replication of IaaS VMs, the vault should be in a different region. This helps ensure the availability of the vault in the event of a disaster affecting Azure datacenters in a specific region.
Tostart ASR configuration for an Azure VM, open the Recovery Services vault ➤ Get started ➤ Site Recovery ➤ Step1: Replicate Application. Select the source, location, and resource group
2. Select the VM to be replicated
3. Select the target region from the drop-down menu
4. Select the target resource group, network, and storage if required. By default, new storage accounts and networks are created for the replicated VMs, but you can customize these settings to select the storage and the network of your choice
5. You can also customize the replication policy settings to change the number of hours for which the recovery point should be retained, or the number of app-consistent snapshots to be taken in an hour Here is some policy for different Replication Policy
6. If you want to replicate and failover multiple VMs, select the option to create a replication group and select the VMs that are part of the replication group. This replication could impact workload performance and should be opted only when a multi-VM consistent snapshot is a requirement
In the subsequent steps, the target resources are created and replication is enabled. Once the replication is completed, the VM is listed at Recovery Services vault ➤ Protected items ➤ Replicated items. From this interface, you can initiate a test failover or planned failovers for the VM to the Azure destination region and resource group. This ensures VM and application resiliency against Azure datacenter region failures.