Virtualization disaster recovery tools ease the barriers surrounding DR, but an effective disaster recovery plan...
also requires tough analytical and technical decisions.
More on virtualization disaster recovery tools
VMware disaster recovery options
Disaster recovery strategies for virtual environments
To put it another way: There isn’t a silver bullet in the form of a disaster recovery (DR) product. Rather, a robust and efficient DR plan requires a wide range of IT efforts, such as deploying multiple virtualization disaster recovery tools, deciding which failures to protect against as well as how to extend services to end users after a disaster.
The DR needs for every organization tend to vary, but all IT shops should evaluate the following virtualization disaster recovery tools and strategies.
Virtualization disaster recovery tools in the market
Major vendors either offer their own virtualization disaster recovery tools (e,g,, VMware’s Site Recovery Manager) or reference architectures that fulfill the need. With Microsoft Hyper-V, for example, you can configure virtual hosts atop multi-site Windows Failover Clusters. And Citrix Systems designed XenServer Pool Replication and other backup and recovery approaches to achieve a similar result.
Third-party vendors, such as Veeam, AppAssure and Acronis, offer a dizzying array of virtualization disaster recovery tools -- many of which focus their attention on augmenting the backup and restore process. These tools create extra copies of data and virtual machines (VMs) that are transferred off site in preparation for a disaster.
NeverFail, Doyenz and Terremark offer products that attempt to reduce the amount of time that passes between a disaster incident and the subsequent return of IT services. These products enable the failing over of entire virtual machines and their workloads. By keeping two copies of every IT service in virtual lockstep, these tools stand prepared to restore services the moment a problem occurs.
Some products from EMC, Dell and HP, among others, focus on storage and the network’s replication of storage from a primary location to the secondary site. As one can imagine, to quickly recover from a disaster, your data should be in two places at once. And the process for transferring that data requires careful management.
Compiling a disaster recovery plan
Despite all of these disaster recovery options, many IT shops find it difficult to point to one and definitively say, “That product does our DR.” When creating a disaster recovery plan, you must determine which disasters to protect against and how quickly your services must come back online. After answering those questions, many IT shops deploy multiple virtualization disaster recovery tools.
Consider the array of failover options that exist to protect VMs from the loss of a single virtual host or an entire rack of servers. Protecting your data center against these disasters may be as simple as implementing your virtual platform’s high-availability features. But even that option requires enough available host resources in reserve for migrating VMs.
High availability also has another weakness: shared storage. Most major virtualization vendors, such as VMware, Citrix and Microsoft offer shared storage-based HA tools designed around a single storage instance. As such, high availability protects VMs after the loss of a host, but it does nothing to protect against a loss of storage.
Safeguarding against the loss of storage is an expensive endeavor. You need to host a second copy of your data at another site, which requires data replication technologies as well as additional storage and networking. Fortunately, good enterprise-grade storage come with a 100% uptime guarantee.
You’ve likely heard that disaster recovery requires up-front planning. More specifically, you must carefully orchestrate the order in which services are brought online to successful carry out a DR plan. The additional steps required to fail back a DR environment into the production data center are just as important. Planning these steps may involve a pencil and paper, but implementing these activities requires software to bring the core services and priority workloads back online.
In the event of a disaster
Finally, some disaster recovery tools connect displaced workers into the DR operations. There are network-centric tools that enable workers to connect via the Internet. You can also deploy desktop-centric products that create a new desktop environment. That environment can then become the interim location where business activities are conducted. How your organization communicates with users will determine the right choice.
A disaster recovery plan once meant supplementing every server with a backup server and ensuring that both machines were separated. But the cost of this practice was untenable for all but the largest enterprises.
Virtualization extends disaster recovery capabilities to smaller business. But, as you can see, disaster recovery is both a technical and analytical exercise for all organizations.