| Server virtualization is still higher on the food chain, but storage virtualization will have its day.
A lot of storage types are put off by the overwhelming hoopla that's been generated by VMware and server virtualization. That's because storage intellectuals feel that it's unfair for server virtualization to get all of this attention considering application and implementation models have been seen in the storage market for years and years. But these people miss the point.
It's all about timing and the IT caste system. Storage consolidation, utilization and automation via virtualization started around 1988 with Veritas Volume Manager, and it's still going on. We went through RAID, LUNs, snaps, continuous data protection, synthetic fulls and dedupe, and we're still not done. Comparatively, virtualization in the server layer had a big-bang arrival. Virtualization in the storage/data layer, meanwhile, has been a slow ride down a lazy river.
The reason storage never received the glory for all of its wonderful virtualization efforts is simple; in the grand scheme of things, the business still views servers as higher on the food chain and storage on the level of primordial ooze. It's an incorrect assessment, but a fact of life. This is because the server layer is the first infrastructure touchpoint associated with an application, which is what the business cares about.
Everyone says "our data is the life-blood of the company," but it's all bull. Generally, business users view data as a part of their application interface; without it, data would be a bunch of 0s and 1s, which isn't all that useful to them. If business people truly believed that all of their shiny applications and gizmos would be entirely useless without data, they would make sure data was stored, protected, managed and delivered using at least a few processes and technologies from the current millennium. To some degree, it's our own fault. We've never been able to easily demonstrate our value in terms that business people understand--we're still geeks in storage. For some reason, it must sound smarter when a server guy talks about efficiency gains enabled by virtualization than when a storage guy says it. Or maybe the storage guy is only saying it to a filing cabinet in the basement? I don't know. What I do know is that everyone is paying attention to server virtualization.
I also know we should stop whining about how our industry was the pioneer of all the stuff that now has our business colleagues excited, and instead be leveraging the movement to our advantage. Consolidation is a good idea for servers and storage. Server virtualization lets us finally get into the '80s, because any big disruptions to our infrastructure allow us (or force us) to take the time to correct legacy issues. Server virtualization will finally enable us to get rid of our direct-attached storage and get our entire data infrastructure networked. It will allow us to re-evaluate our current, dumb ways of doing backup and do disaster recovery even if we're only a 20-person company. Server virtualization can have the glory, as long as it gets the core data architecture where it needs to be.
I don't care if the business guys think Microsoft invented "system" virtualization or that it was bestowed upon us by aliens three months ago ... this is an opportunity for storage folks to evolve out of the sludge. Who knows, maybe next year we'll even get invited to the holiday party.