Virtualization… everyone is doing it. Virtualization is certainly in vogue; in fact, Gartner says it now has close to 50% penetration in the enterprise. And with benefits like cost savings and increased efficiency, who wouldn’t want to take advantage?
When a hot new technology touting such great benefits emerges, it can be tempting to jump right out of the gate without carefully considering the requirements to deploy it within your organization. But this can be dangerous to your business. When making the decision to virtualize, don’t do it with blinders on. There are several things to consider when transitioning from physical to virtual, especially as it relates to disaster recovery. Let’s take a look at four of them.
1) Plan your resources effectively. Application performance can suffer if any one resource is over-allocated. This can affect your DR plans as well. Bottlenecks arise from many different sources, including CPU, disk I/O, RAM and bandwidth. Therefore, if you’re going to virtualize, you should make sure that you have the capacity and resources to do it without risking the degradation of your resources.
2) Plan for failures. In the event of a disaster, the speed with which you recover depends not only on the recovery tools that you’ve put in place, but also on the failures for which you’ve planned. Perhaps you’ve planned for a large-scale disaster, but haven’t given consideration to an application being down due to planned maintenance. It’s important to note that not all disaster recovery solutions protect you from all sources of downtime. Don’t let the ease of VM image mobility trick you into a false sense of security.
3) Evaluate dependencies. Applications rarely exist on single servers. More often, there are interdependencies between servers that need to be considered. Be sure to map out all components of an application or service, and make sure that all parts are equally protected in a disaster. They should – at the very least – be protected enough to meet your agreed SLAs.
4) Understand requirements. Virtualization tools promise many capabilities, and you need to make sure you have the resources to take advantage of them. For example, planning DR around moving images, and then discovering that you’re not able to move images quickly enough really throws a wrench into your DR plans. Make sure you understand exactly what is required of any solution, and ensure that you are able to meet those requirements.
When done correctly, virtualization promises great benefits and return on investment. However, when done incorrectly, it can have a tremendously negative impact on your business. Disaster recovery planning for a virtualized world is no different (see Five things that out-of-the-box DR tools can’t provide). Just like with anything new, successfully running a virtual environment just takes planning, a bit of patience, and having the right tools to do the job.
Server image by renjith krishnan