Storage virtualization is really nothing more than taking multiple storage devices, and abstracting them so that the end user can more easily manage a larger amount of storage without having to deal with the complexities of multiple devices. Storage from multiple devices are combined and appear as a single resource. When a request for data is made by a user, the request is made to the virtualized environment, which then passes it on to the specific device on which the data is held. The end-user doesn’t need to know the physical location of the data; the virtualization platform takes care of that. Usually, the extra step involved doesn’t introduce any noticeable latency.
Even smaller organizations are starting to deploy virtualization solutions, and it makes sense for many of them today, especially from a scalability standpoint. For a small organization, adding extra storage capacity when using a direct storage model involves an additional step, and smaller companies may not have the time to handle it—the virtualized model makes adding extra storage a very straightforward process that can be done in real time and without taking down the system.
So let’s take a moment to look at what’s involved.
The advantages are obvious from the beginning; besides simpler usage, it allows for greater and more efficient use of all storage devices. From the admin’s point of view, a big advantage is easier management, since it allows all storage devices to be managed at once, from a single GUI. This makes it much easier to see storage trends. Disaster recovery is also enhanced through virtualization; since this is typically addressed on a physical-to-virtual basis. That is, it is no longer necessary to back up a physical server to another distinct physical server at the backup site, which saves both time and money. On the back end, when disaster strikes and recovery becomes necessary, it’s naturally a lot quicker to recover data from a single virtual device than it would be to recover it from multiple, distinct devices.
In a virtualized environment, storage requirements are lowered because of thin provisioning, which eliminates the need for allocated and unused storage, and greatly increases storage utilization rates. There really are very few technological challenges to implementing virtualization in a storage environment; it’s deceptively simple. Most of the work is done in the planning stage rather than the implementation stage; once you’ve done your homework, you can be up and running in a day. The planning stage may involve several things, including looking at the requirements of each application, the requirements of the hardware, and your requirements for disaster recovery, business continuity, and high availability.