Managing and protecting all enterprise data


Evaluate Weigh the pros and cons of technologies, products and projects you are considering.

Five things that should happen in the data storage market in 2014

It's a new year and I'm newly optimistic again, hoping that 2014 brings a healthy dose of clarity and reality back to the data storage market.

It's a new year and Editorial Director Rich Castagna is newly optimistic again and hoping that 2014 brings a healthy dose of clarity and reality back to the data storage market.

I don't know what constitutes a tradition, but in these days of instant gratification, I think doing something two years in a row qualifies. And since I wrote a similar column last year at this time, I will now present my traditional New Year's column on five things that should happen in storage in 2014 but probably won't.

Before I issue my 2014 tirade, here's a quick review of my wish list from the first annual should-happen-in-the-data-storage-market-but-won't column.

Last year I hoped for cloud storage standards (still hoping); data classification to come back into vogue (I'm still optimistic); primary storage dedupe (saw some progress); some reality-based ROI/TCO talk about virtualization (ha!); and real alternatives to RAID (can you say "erasure codes?").

In retrospect, that was a pretty modest wish list, and the potential was there for it all to happen. In a perfect world, it would have. But we don't live in a perfect world, so here I go again.

WISH 1: Will someone please define software-defined storage? According to my latest count, a million vendors (out of a million and one) are selling something they're calling software-defined storage. I'm sure they all feel very good about having a product in this new, hotly contested market -- and they'd feel even better if they only knew what the heck software-defined storage meant. It's been bandied about so much that it even has its own acronym: SDS (apologies to the 1960s Students for a Democratic Society; my, how an acronym can change in a mere 50 years).

Put the names aside for a moment and just compare some of these products to see how all over the map they are. There are real software-based storage controllers capable of supporting limited amounts of storage capacity in a DIY array if you really want to scrape your knuckles and maybe lose all those tiny screws trying to put one together yourself. Then there's the software-defined storage that doesn't actually do any defining, but rather kind of redefines things and pools existing storage systems into one giant storage resource that can be managed from a single window. If that sounds like good ol' storage virtualization, it's because it is. And then there's the kind of software-defined storage that sort of falls in between and if your rep convinces you to buy it, you'll spend the next year trying to figure out exactly what it does.

WISH 2: Flash! We need some clarity about solid-state storage. Flash is cool, quick and confusing as hell. We're a few years into the "flash in the enterprise era" and you'd think all that solid-state stuff would be more mainstream -- you know, just another souped-up drive. But more and more alternatives related to form factors, placement in the storage environment and various functions are upping the confusion quotient for storage managers trying to spend their limited budgets wisely. True, it's hard to blame a technology that's just evolving real fast, but flash's growing complexity is being aided and abetted by vendors who don't want users to be able to compare their products all that easily.

Let's face it; all-flash arrays are still pretty exotic and should probably only be considered for apps that need to go from zero to 60 in a blink of an eye. More than the performance of a little solid-state tucked into an otherwise spinning-disk array can offer -- but that setup can be shared among a lot of applications pretty effectively. And dedicated solid-state storage installed directly in a server is the easiest, most expedient way to expedite a performance-hungry app. Now that I've cleared up all that solid-state confusion, vendors can do it, too.

WISH 3: Limit vendors to only two buzzwords per product announcement. Your product can't possibly combine big data, virtualization, software-defined storage AND the cloud. Believe the hype and you'll believe a plain old storage array is some kind of super box that can crunch through billions of bytes in a nanosecond, is so virtualized and software-defined that it's invisible, and is capable of turning your stubbornly anchored-to-the-ground data center into a fluffy cloud. I guess it's pretty easy for vendors to make those claims since all those tags defy anything remotely resembling a clear definition. But if you find yourself buying into that stuff, and you're signing on the dotted line for one of those Swiss Army knives, you may also be interested in some swampland I'm selling.

WISH 4: Let's put the "tape is dead" talk to rest and just admit that "tape is alive and well." Tape has died or teetered on the brink of death more times than Generalissimo Franco. And each time, after it's been declared dead, some smart aleck company like Oracle, IBM, Spectra Logic, HP or Quantum comes out with some new tape technology that makes tape the highest capacity media bucket around, has it operate fast enough to give disks a run for their money, or adds capabilities like LTFS that could expand its role in the enterprise and make it a cheap alternative to spinning disk. That seems pretty lively for a "dead" technology.

WISH 5: Big backup app vendors will wake up to the perils of BYOD and provide some useful data protection tools. Here's my conspiracy theory: Every backup vendor and their uncle has a well-crafted, easy to implement and even easier to use mobile device backup app tucked away in a vault in an old salt mine in Nevada. They'd try to sell you an endpoint backup app right now, but they know you're not interested. And just how uninterested are you? Our survey data shows only 4% of you are actually backing up smartphones and tablets. A whopping 69% don't do anything about those devices, and approximately 20% have end users do their own backups (now there's an effective data protection policy). So backup vendors aren't going to waste their marketing bucks just yet; they'll wait until BYOD results in massive data loss for some unlucky company and turns it into bring your own disaster (BYO{kaboom!}). That ought to get someone's attention.

Happy New Year, everyone; I hope your wishes come true, too.

About the author: 
Rich Castagna is editorial director of TechTarget's Storage Media Group.

Article 5 of 8

Dig Deeper on Storage vendors

Join the conversation


Send me notifications when other members comment.

Please create a username to comment.


Some great comments! Entertaining and spot-on. I particularly agree with you on #4 and #5.

#4 - No, tape isn't dead, and with all the focus on disk backup (certainly the preferred first landing zone) users sometimes neglect to consider tape functionality in their backup products. A simple "move this stuff to tape" feature isn't really enough. You have to be able to manage and track the process.

#5 - Agree completely here, but I don't want to turn this into a vendor commercial. All I'll say is that you CAN actually find some solid desktop/laptop/mobile protection solutions out there, and the good ones are pretty simple to use. Though from an enterprise perspective you definitely want something where IT can lock in the policies, because as you note with tongue firmly in cheek, end users doing their own backup is "an effective data protection policy." Lol!

Happy New Year!

Peter Eicher
So how about a followup to this? How well did vendors do in 2014, and what are your hopes for 2015?
Sharon -- you were one step ahead of us. But it looks like some of our wishes (flash, software-defined storage) were granted. For the most part, at least.

Get More Storage

Access to all of our back issues View All