Right now, controlling storage costs is a big deal in most shops. There are a number of reasons for this including the fact that it’s just such a big part of the IT budget these days, and getting
bigger all the time. It’s also the fact that the hardware isn’t what’s growing the fastest, it’s a management costs. But, I think that the way that most shops are looking to resolve this isn’t by going back to buying everything from a single vendor, though we may see some of that in the short term. I think that most shops are looking at storage virtualization to address the problem.

Storage virtualization addresses the business problems we have with storage in a couple of ways.

1. It will drive down the cost of physical storage by making storage even more of a commodity than it is today.

2. It will create a single point of management for all of the different kinds of storage, thus reducing the cost of management.

So, as you can see, storage virtualization attacks both the issues, the cost of the hardware, as well as the management costs.

However, care will need to be exercised by anyone heading into the storage virualization world. There are all kinds of possible stumbling blocks out there for the unwary storage team trying to implement storage virualization. For example, if you are currently using replication, you need to move away from array based replication. Why? Simple, you don’t want to have to know/manage the array, so the replication needs to be part of the virtualization engine, or some third party replication tool that will work with your virtualization engine, otherwise, you will still need to understand the array based tools and you then lose some of point #2 above.

That’s just one example of the potential pitfalls involved. So, storage virtualization is, in my opinion, the future solution to some of the business problems we face with the implementation of storage but finding the right solution is going to take some time, and some careful planning.

–joerg