Over the last several weeks this blog has featured a series of posts about the benefits of the Hybrid Cloud– and today marks a major Hybrid Cloud milestone.
I am excited to announce the General Availability (GA) of Windows Azure Hyper-V Recovery Manager service.
Windows Azure Hyper-V Recovery Manager Service protects your on-prem applications by orchestrating the protection and recovery of Hyper-V Virtual Machines running in a private cloud (i.e. System Center Virtual Machine Manager 2012 R2 or System Center Virtual Machine Manager 2012 SP1) to a secondary location.
Over the years, as I have spoken with the VMWare community, I have heard things like this: “Well, with Windows Server 2008 you did not have Live Migration; let me know when you have that.” Recently, that one missing scenario has been SRM. Well here it is! And I want to emphasize that this solution is so much easier to use and the way we have architected it is a much more modern, cloud-centric way of doing things.
Hyper-V Recovery Manager assembles some core elements of our Cloud OS strategy (Windows Server 2012 Hyper-V Replica, System Center Virtual Machine Manager, and Windows Azure) to deliver a cloud integrated Disaster Recovery Solution. Reaching GA means that the service is now backed by support and SLA assurance, and IT administrators can use it in production environments. We have had a number of customers running this in production in preview, and their feedback has been straightforward: The solution is incredible and it is ready for general availability.
While building Hyper-V Recovery Manager we listened carefully to customers all over the world – and a consistent piece of feedback was that although they had a DR strategy in place for their business, they were unable to implement it for all their applications. On top of this, Disaster Recovery (DR) solutions have historically been expensive and complex to deploy. These factors often leave IT Pros in a situation where they implement DR protocols for their critical applications but leave many other parts of their business exposed.
In one of our reviews six months ago, I asked the team to walk me though – side-by-side – what it was like to setup DR for 500 VMs using Hyper-V Recovery Manager compared to our primary competition. I literally counted the number of mouse clicks as one of the ways to understand which solution was more complex. To my surprise, the team demonstrated they could do this in less 20 clicks of the mouse! By comparison, I would have needed to come back the next day before it could be setup with the competition. More on this later.
Hyper-V Recovery Manager’s singular focus is to deliver a simple and cost-effective solution. By leveraging the power and reach of Windows Azure and Windows Server’s in-box replication technology in Hyper-V Replica, Hyper-V Recovery Manager can provide a high-performance, low-cost DR solution for every inch of an organization.
The diagram below captures the high level architecture of the service. The service itself is hosted in Windows Azure and it securely communicates with System Center Virtual Machine Manager via a lightweight provider installed on the VMM Server. VMM servers send the metadata about Virtual Machines to the service, which then uses it to orchestrate the protection and recovery of Hyper-V Virtual Machines in the private cloud. On-going VM replications happen site-to-site via Hyper-V Replica.
I really want to emphasize just how unique this is compared to our competition. The control plane for DR is now in the cloud (running as an Azure SaaS app). The VMs being replicated between clouds never come to Azure – Azure is just the control plane where the policies, automation and status are stored. This use of the public cloud is a game changer. Windows Azure offers a global, highly available cloud platform that is available anytime and anywhere. By leveraging Azure to offer a DR management service, we eliminate the need for our customers to invest the time and resources to build a high-availability platform for their DR needs.
I believe that this model of the control plane being in Azure as a highly available SaaS app is the architecture we should all be leaning towards.
Key Differentiators
Hyper-V Recovery Manager is unlike any other DR solutions previously available. A few key elements that really set it apart include:
- Simple Setup and Configuration
Hyper-V Recovery Manager has been designed for simplicity. Since it’s a hosted service, customers don’t need to install/deploy additional servers or worry about servicing/patching. By installing a lightweight VMM provider, they can extend their existing VMM environments and quickly make it DR ready. By leveraging VMM’s management capabilities, the service can also push cross-site configuration across a large number of Hyper-V hosts and Virtual Machines. This enables scale without comprising on simplicity. You also setup your DR policies at the cloud level. This allows you to express policy such as, “I want all VMs in the Finance cloud to be replicated to my local service provider,” or, “I want all VMs in the Redmond Cloud to be replicated to the Chicago Cloud.” - Automated Protection
Hyper-V Recovery Manager leverages the capabilities of Hyper-V Replica to provide on-going replication of Virtual machines. The service also monitors for changes like VM migration, additions of new clusters, changes in network setting – and it automatically re-configures VM protection if needed. This ensures that Virtual Machines remain protected throughout the lifecycle without manual intervention. - Remote Monitoring
Hyper-V Recovery Manager can be accessed via the Windows Azure Portal from any internet-enabled device. VM protection status can also be remotely monitored and, in the event of an unforeseen disaster, customers can begin the recovery operations remotely. - Orchestrated Recovery
Cloud-based Recovery Plans enables automated DR orchestration by sequencing the failover of different groups of Virtual Machines. For example: In a multi-tiered application you can arrange the Database tier, middle tier, and web tier in different groups within a Recovery Plan and sequence it such that the database tier fails over and boots up before the other tiers. Additionally, recovery plans can be customized with scripts and manual actions. For example, one can combine a SQL Always On failover PowerShell script with a Hyper-V Recovery Manager Recovery Plan for application-specific DR operations.
Scenarios and Use Cases
Enterprise customers can leverage Hyper-V Recovery Manager in their existing VMM-based private clouds to quickly implement a DR plan for their virtualized workloads and applications that are compatible with Hyper-V Replica. To get an idea of what these kinds of scenarios look like, consider these use cases:
- Failover and Recovery in case of an outage/natural disaster
In situations like these, customers can use Hyper-V Recovery Manager to bring up their applications on a secondary site and achieve business continuity. In scenarios where early warning is available, customers can perform a Planned Failover, which will ensure zero data loss and achieve disaster avoidance. In scenarios where events happen suddenly, customers can perform an Unplanned Failover. Using Hyper-V Replica’s near-sync replication feature, customers can achieve an RPO of as low as 30 seconds when failing over. Incredible, right? - Planned Maintenance with Minimal downtime
Periodically, datacenters need to take planned downtime for various servicing activities, e.g. AC maintenance or routine servicing. During a planned maintenance, administrators can use Hyper-V Recovery Manager’s planned failover capability to bring up the applications on the secondary site quickly. Once the servicing is complete, they can use Planned Failback capability to bring the application back up on the primary site. - Disaster Recovery Drills with no downtime
In order to maintain regulatory compliance, businesses need to demonstrate DR readiness and perform drills on a periodic basis. Hyper-V Recovery Manager’s Test Failover capability enables customers to perform a DR drill on an application without impacting the performance of the production/primary site. In this scenario, a test copy of the VM is brought up on the secondary site which can be tested to ensure correctness without any impact on the production workload. - Planned Datacenter Migrations
Customers such as United Airlines are using Hyper-V Recovery Manager to migrate workloads from one datacenter/location to another as part of their datacenter consolidation. - Prove Your DR Scenario for Compliance
One of the things we constantly hear is the concern from customers that they are not 100% confident that the DR and failover will work when they need it. Clearly this is a problem. It goes without saying that the last thing you want to be doing during a crisis is seeing whether or not your DR strategy actually is functional. With Hyper-V Recovery Manager it is incredibly simple to do a Planned Failover and demonstrate/prove to that the DR strategy works. This is critical for compliance, business contingency planning, and the peace of mind to sleep at night! - Hosters and Service Providers can leverage Hyper-V Recovery Manager to offer managed DR services for tenant Virtual Machines hosted in their environments. They can also build new offerings around protected VM, Recovery Plans, and perform DR operations on behalf of their customers in case of an event or as part of a DR drill.
We have worked closely with customers to refine this service, and I think the enterprise-ready, battle-tested nature of this GA will really show. These are high-value scenarios, and I this feedback has been consistent from the customers that have already been using it.
There are already some great examples of Hyper-V Recovery Manager-based Disaster Recovery solutions at work for customers like Aston Martin, United Airlines, Dartmouth-Hitchcock Medical Center, and Pošta Slovenije.
With general availability, I look forward to more evaluations and feedback about this service and our overarching cloud service model. Our goal is to study the feedback we receive and identify the ongoing improvements we can add to the service.
I encourage you to visit the Windows Azure site for more information on Hyper-V Recovery Manager, and you can also check out additional product documentation here. There’s also a Hyper-V Recovery Manager forum on MSDN for additional information and easy engagement with other customers.
Again, I’d like to congratulate the Hyper-V Recovery Manager on a great release, and I want to encourage all our partners and customers to make the most of these DR solutions!
To learn more about Hyper-V Recovery Manager, check out the preview announcement and this overview of cloud-integrated disaster recovery from the “What’s New in 2012 R2” series.
Honestly my biggest ask is this: Just go try it! Especially while the price is 50% off through March 1st!
To learn more about how to purchase Hyper-V Recovery Manager, and get into the details about pricing options, visit this page on the Windows Azure website.