In case you didn't notice, storage virtualization solutions have become a commodity.

In the past month, I've gotten no fewer than six emails from storage optimizations companies talking about how they're bringing something to market or pointing out a solution that we haven't talked about.

In the past month, I’ve gotten no fewer than six emails from storage optimizations companies talking about how they’re bringing something to market or pointing out a solution that we haven’t talked about. This is amazing to me because at the end of 2012, we could almost count on one hand the number of storage optimization solutions with VDI-specific applications (Atlantis, Fusion-IO, Nutanix, Virsto, DataCore, Tegile, X-IO). Since then the market exploded en route to what I believe is the commoditization of storage optimization.

Over the past few years, we’ve seen many other companies join the usual suspects in the public storage optimization consciousness. XtremIO, which sponsored BriForum one year while still in stealth mode, never made it to launch before being acquired by EMC. They're credited with setting the parameters for the perfect VDI storage solution: block level, single instance storage where the optimization is performed before the data is committed to disk, not after. Since then, we've added Nimble Storage, GreenBytes, PureStorage, Simplivity, Tintri, Whiptail (which was just acquired by Cisco), and Nexenta to the list of companies with desktop virtualization-specific solutions. Then over the course of the summer, more companies appeared on our radar, like Velobit and Infinio. (This is by no means a complete list)

Even software companies are getting into the game. Liquidware Labs, a software company known for user environment management and OS migrations, has created FlexIO, which is included as part of their ProfileUnity product. Dell’s vWorkspace group has been adding optimizations for deploying and executing VMs to Hyper-V, VMware bought Virsto, and Citrix added IntelliCache to XenServer. 

This article alone will probably result in more emails that say “Hey, you didn’t mention us,” which will serve to reinforce two points. The first is that storage optimization, through any of several different methods, is accessible. No two solutions are exactly the same, although you don’t have to look too hard to find similarities. Some use memory and/or single instance storage, others focus on tiering data across various mixes of fast (expensive) and slow (cheap) storage mediums, while others are 100% flash-based. Hell, at VMworld this year, we found that the solutions from Atlantis and Greenbytes were so similar that we couldn’t choose one over the other for the Best of VMworld award in the Desktop Virtualization category.

The other point that is reinforced by all the storage optimization solutions on the market is that VDI storage has, for the most part, been solved. Today, the challenge is not that VDI storage is too hard or too expensive; rather, it’s deciding which solution to use of the twenty or so that are out there. In no way is any of this a bad thing. Remember, just a few years ago, we were faced with the decision of trying to shoehorn our virtual desktops into existing SAN or to build an entirely new one specced out for desktops. The kids have it so easy these days.

So here’s to the commoditization of storage optimization! Are some solutions better than others? For sure, but with the varying use cases out there, what’s good for me might not be good for you. The point is, there are solutions for everyone now, and we don’t have to mortgage the datacenter to get them.

Join the conversation

1 comment

Send me notifications when other members comment.

Please create a username to comment.

I agree that storage has become more accessible. But for me it's been like that for at least 3 years, having been designing and selling solutions. The real difference is confidence and market awareness amongst people. What was once specialist and niche (e.g Whiptail) now has a badge on it. Almost all the big names have something.

What I am talking to people about today is gradually becoming less about the kit and far more about the management ongoing.

Do you need non-persistent desktops if the storage is doing the clever stuff? e.g Dedupe

How will the desktops be provisioned and maintained and updated depending on persistent vs. non.

How well does your storage option integrate into View, XenDesktop (PCS/MCS) if at all? This is not a simple answer in most cases!

Do you want a storage model that grows in 1,500 desktop increments or 10 at a time? This might be based on VDI capability or challenges around who and how desktops are paid for.

Who will support the SAN and Hypervisor? Which of course leads on to possibly the next commodity of the future... Converged solutions from the like of Nutanix, Pivot3 and newcomer SimpliVity.

Hyperconvergence... Now I wonder if it will ever become a commodity.