Azure Site Recovery

Overview

Microsoft introduced the Azure Site Recovery (ASR) service in public preview in June 2014. ASR initially focused on Hyper-V–to–Hyper-V and on-premises Hyper-V–to–Azure recovery scenarios. Over time, the service evolved to include additional capabilities, such as support for various operating systems, Azure IaaS VMs, and complex workloads that you could replicate in coordination with each other so you could bring them online in Azure with a similar recovery point. The simplicity, stability, and cost effectiveness provided by ASR has made it a very popular service used by organizations for their business continuity and disaster recovery (BCDR) strategy.

The key features of this service include the following:

• Simplified interface to set up, monitor, and manage BCDR ASR provides an intuitive interface that makes it possible to easily set up, monitor, and manage the service.

• Support for most commonly used operating systems ASR supports the replication of most operating systems, and the support list is expanding on a regular basis. This makes it possible to use ACR as a one-stop BCDR solution for most organizations.

• Support to define replication schedules to meet the RTO and RPO objectives for most organizations ASR supports replication as low as 30 seconds for Hyper-V VMs and continuous replication for Azure VMs. VMware support is currently under preview but is expected to allow continuous replication.

• Support for any workload ASR supports the replication of any workload as long as it is hosted on a supported operating system. This makes it a viable solution to address most organizations’ needs.

• On-premises Hyper-V–to–Hyper-V replication, monitoring, and recovery orchestration across multiple DR sites ASR supports replication monitoring and failover coordination for on-premises Hyper-V–to–Hyper-V servers across multiple interconnected sites. Hyper-V replication is used for data replication, but ASR helps automate all the recovery steps from the cloud.

• Azure–to–Azure IaaS VM replication, monitoring, and recovery orchestration across Azure regions ASR added support for Azure VM replication and failover across Azure regions in 2017. This allowed organizations to set up a BCDR strategy for their critical cloud workloads across multiple Azure regions.

• On-premises Hyper-V–and physical servers–to–Azure replication, monitoring, and recovery orchestration ASR supports the entire recovery management for workloads hosted in on-premises Hyper-V and physical servers to Azure VMs. Azure Storage hosts the replicated data. VMs are created only after a failover is initiated, reducing the costs associated with a secondary datacenter.

• App-consistent snapshots to recover applications more efficiently ASR supports using app-consistent snapshots for replication to ensure that applications are replicated using disk data, all data in memory, and all transactions in process.

• Simplified and cost-effective pricing ASR charges a fixed fee per server for the replication software; Azure Storage costs for replicated data are charged only after the VMs are failed over in Azure. This makes it extremely cost-effective when the DR site in Azure is not actively used to host VMs, as the majority of costs are related to running the VM. All this makes it possible for most organizations to afford a BCDR solution for their environment.

• Testing DR without interruption to production ASR supports the activation of the replicated data in isolated networks in Azure. This enables you to test VMs to make sure your application, database, and other workloads are working as needed before you need to actually use them in a DR scenario. This also enables scenarios in which application upgrades can be tested in the Azure cloud before implementing them in your on-premises environment.

• Integration with SQL Server AlwaysOn ASR supports integration with SQL Server AlwaysOn to allow for seamless recovery of both interconnected application and database workloads.

• Integration with Azure Automation Azure Automation enables you to set up scripts and automated actions to provision other Azure services or run pre- and post-failover scripts as part of your automated recovery procedure.

• Multi-VM consistency using replication groups ASR supports setting up multi-VM replication groups so that multiple VMs are replicated together, and app-consistent and crash-consistent recovery points are created to facilitate failover. This enables you to address scenarios that require multiple VMs to be maintained at the same consistency.

ASR supports various recovery scenarios that can be used by organizations in different ways, depending on their individual needs. The following sections cover the two most important scenarios that ASR supports: Azure–to–Azure disaster recovery and on-premises Hyper-V–to–Azure disaster recovery. Read these sections to obtain a better understanding of how both these scenarios can be set up, managed, and monitored using ASR.