You might be tired of hearing it, but almost every organization that suffers a major event without a disaster recovery (DR) solution in place will fail to get back on its feet. So if you want to make sure your business survives -- and you should -- then you need to think about enacting disaster recovery strategies.
A full-fledged DR infrastructure is often out of reach for midmarket companies, which is why they tend to leave DR to the last minute. Let's get this straight: Disaster recovery no longer consists of just taking backups and storing them off-site. Disaster recovery strategies that rely solely on backups will often fail simply because they aren't put to the test on a regular basis. Too many organizations have discovered that the backups they were taking didn't work to recover data and systems after a disaster.
With the advent of virtualization -- especially server virtualization -- the nature of DR has changed, and companies can now rely on a cloud-based infrastructure to provide business continuity in the event of a significant failure. The disaster recovery strategy is simple: implement a server virtualization infrastructure internally and acquire appropriate cloud-based infrastructure extensions to support DR. There are lots of providers, making cloud-based DR pricing very affordable. Companies such as SunGard Data Systems Inc., GoGrid, Microsoft, RedPlaid Managed Hosting and many, many more offer virtual server infrastructure extensions that can take on your workloads when significant events prevent you from operating your services internally.
While virtualization does make cloud DR simpler, you are still required to perform your due diligence when selecting a cloud DR provider. Use the following guidelines:
- When implementing server virtualization, try to focus on a single virtualization provider.
Mixed virtualization infrastructures are more difficult to manage and protect from disasters.
- Make sure the cloud provider's virtualization infrastructure offers the same type of
environment as your own. While many vendors offer support for several different virtualization
solutions, you'll find that most tend to have advanced expertise in only one. Make sure this is the
one you are using..
- Determine which type of DR extension you require. Cloud DR providers offer several different
types of services, from self-service virtual infrastructures to complete hosting services. Choose
the one that best fits your management model..
While virtualization does make cloud DR simpler, you are still required to perform your due diligence when selecting a cloud DR provider.
Trusting your core systems to cloud DR providers can be risky. This is why you need to make sure
your critical data is constantly protected and secured.
- Start small. Start with one service and then, when you're satisfied the service operates
properly both internally and externally, move on to other parts of your infrastructure.
- Perform regular recovery tests. If your infrastructure is to be protected by a cloud DR system,
then you must make sure that it works during a disaster. Simulate as many possible disasters as you
- Choose your replication engine carefully. Part of the magic in cloud-based DR through server
virtualization is that your core services now run on machines that are nothing but a set of files
in a folder. To create cloud
disaster recovery strategies, you simply need to replicate these files to a service provider
and boot them up in the event of a failure. Make sure the replication solution you choose can
support open file replication so that your DR copies will be exact duplicates of your running
These seven caveats will help ensure that the foundation of your new cloud disaster strategies will be stable and will provide long-term service for your organization. That doesn't mean you should put all your eggs in the same basket. Backups -- internal and external -- are still a good idea. Make sure your disaster recovery strategy will include a complete set of services, both traditional and virtual, and that your IT staff fully understands how to react when something goes awry. Protection is a must for any organization. Perfect protection may be impossible, but if you approach it the right way, cloud-based DR may be the best thing that ever happened to your organization's need for continuity.
Danielle and Nelson Ruest are IT experts focused on virtualization, continuous service availability and infrastructure optimization. They have written multiple books, including Virtualization: A Beginner's Guide for McGraw-Hill Osborne, and MCTS Self-Paced Training Kit (Exam 70-652): Configuring Windows Server Virtualization with Hyper-V for Microsoft Press. Contact them at firstname.lastname@example.org or email@example.com.
This was first published in September 2010