A recent survey from Spiceworks reports that virtualization is the top IT initiative among SMBs. The survey goes on to say that 67% of SMBs are planning to move core applications to virtualized environments and that 46% have adopted some form of cloud services. These figures represent a healthy increase over results from a similar survey conducted a year earlier.
This adoption trend is consistent with that seen in Axcient’s growing customer base. Server consolidation is a driving factor, since it reduces hardware and energy costs, speeds server deployment, and simplifies administration. But what about ensuring the uptime of a business that’s dependent on those consolidated, servers?
An increasing number of SMBs are cleverly employing virtualization as a way to affordably protect their servers, while also keeping recovery time objectives to an absolute minimum. The approach entails taking periodic, incremental snapshots of the live virtualized servers. Then, if a server fails, it’s virtualized on standby hardware so that business operations can resume. So, how does one achieve this “virtual reality”?
There are backup tools that utilize APIs provided by the popular hypervisors. However, these APIs are proprietary and lock a backup or failover mechanism into a specific hypervisor implementation. The issue is exacerbated when adding cloud to the picture for disaster recovery. In other words, you need to consider whether all the servers, regardless of whether they are physical or virtual, can be brought back to life as a cohesive virtual office in the cloud.
An alternative approach utilizes standard mechanisms for server protection, which affords tremendous flexibility. For example, VSS snapshots on Windows servers are identical, regardless of whether you’re running Windows on a bare metal server or as a guest operating system on VMware, Hyper-V, or Xen. This opens up several very useful possibilities:
- A single data protection solution can protect a heterogeneous environment consisting of both physical and virtual servers. You can even protect virtual servers hosted on dissimilar hypervisors.
- Switching from one hypervisor implementation to another, whether it’s for business or technical reasons, is simple. Just use a backup from the first hypervisor to perform a bare metal restore to the second one. As an added bonus, you don’t lose any of your recovery points from when the server was hosted on the first hypervisor.
- Migration from a physical to virtual environment is straightforward. Similar to the point above, you can use a backup from the physical environment to perform a bare metal restore to the virtual environment and still retain all your recovery points. Going the other direction, from virtual to physical, is also possible if you decide later to dedicate all the hardware horsepower to a single server.
This approach ensures uptime protection for an SMB, irrespective of changes in virtual or physical server technologies. For demonstrable examples, check out the Foritfy and Cloud Continuity offerings by Axcient.
So, go ahead and choose your virtual reality with confidence. You can always alter it in the future.