The (storage) times they are a-changin'​

As someone who’s worked in the industry for 30 years now, one thing that I’ve become used to and actually embrace is change. Whether it be a move to solution-oriented selling or a disruptive technology emergence, change stops me from getting tired of the same-old way of working and reminds me of how I could never have survived in a 9-5 office job with no variety.

If we rewind a few years, the storage market was actually getting quite dull and mundane. The big five vendors were monopolising a great deal of the market share. Customers were using “big-iron” storage for their tier 1 requirements, vanilla mid-range storage for tier 2 and way too many consultants were sketching “ILM” pyramids and over-charging for borrowing your watch and telling you the time.

Along came flash media, in-line deduplication and a whole bunch of upstarts and suddenly things became exciting again. Just a walk around VMworld counting the number of storage vendors present showed us that things were changing and the dominance of the big guys was finally coming to an end. Of course, with the arrival of a bunch of new ways of solving problems comes a degree of market confusion but this is where the channel could once again add value and help its customers wade through the fog.

Now, with the Dell-EMC merger moving along and the (partial) success of hyper-convergence in the market, we’re finally seeing the long-lauded commoditization of storage. It happened such a long time ago with x86 and networking that I was beginning to question whether it would really happen for storage, but it’s now happening.

So how come we’re still seeing a whole industry full of mid-range storage products comprised of clever software sitting on industry-standard, OEM servers with run-of-the-mill disk shelves. The only differentiation between these being a nice logo and some sales people with funky-colored sneakers telling you how differentiated their platform is? It still comes back to the age-old “buy vs build” argument that’s been around for years.

Would you rather buy the clever software, install it onto your platform of choice and then take ownership of the integration points, the multiple service contracts and the application integration? Or would you rather buy an all-in-one sealed unit with single service? Now I’m well aware that there’s pros and cons on both sides of this argument, however I’ve often wondered why there’s never really been that much of a sensible alternative out there?

A long time ago, a wise sales guy said to me “no one buys storage”. With me looking at him confused, he continued “people buy solutions and storage is just a necessity to deliver the solution – find out why they want to buy x terabytes of storage and what they plan to use it for – then focus on that requirement, not the just the capacity”. He was right of course but still now in 2017, I hear storage vendors talking about the media and not the solution impact – something which never ceases to amaze me.

Which brings me something I call the storage platform squeeze.  Some years ago we saw the emergence of the term “Software-Defined Storage” (SDS). I must confess it took a little while to get my head around it but essentially it was the abstraction of the control plane from the data plane and was around using software installed on non-proprietary, industry-standard hardware to manage “dumb” storage whose job it is to protect data and serve it in a timely manner. The problem was that for this model to exist, there had to be reliable, performant commodity storage tin available that could provide simple integration points to the software. In short, that existed but was hard to find and even then, it was often painful to deploy. Add to this the absolute abuse of the term “Software-Defined” by marketing departments and you had the majority of storage buyers and architects retreating to their comfort zone and buying from the aforementioned big five.

Let’s jump again back to 2017 and look what we have now. Without a doubt, storage hardware has improved both in terms of reliability and predictable performance. No one has to study the failure rates of individual drives as they used to back in the early ‘00s and flash has helped clear some of the bottlenecks that made clever-caching storage arrays an absolute must-have for high throughput environments. We’re also starting to see drive and component manufacturers move into the enterprise space with customers realizing that the high margins demanded by many storage array manufacturers and the buy vs build dynamic is changing. Add to this the arrival of public cloud offerings and you have a scenario where more options are available to deliver a solution requirement that just needs storage as a necessary evil on the back-end.

What has been happening quietly in the background is interesting. FalconStor is a great example of a vendor who’s been quietly beavering away solving data management challenges for their customers in a wide-variety of fashions that have been labelled SDS – something that’s been both a blessing and a curse. It’s helped customers who want SDS to find them but at the same time it’s confused people who haven’t been able to extract the four key impacts that the FreeStor product can provide – migration, continuous availability, disaster recovery and analytics, but crucially across multi-vendor implementations with a common management framework.

If you take FalconStor’s FreeStor product and re-analyze the buy vs build argument, then there starts to emerge a compelling argument to reassess the gains that can be made from this type of implementation. The good news for those interested in taking advantage of FreeStor is that, due to its pedigree and customer implementations worldwide, there’s some great case studies available that can be exampled and provide the foundations for such an analysis.

In summary, storage is subtly going through another change. For someone who’s been selling storage systems for a long time, it’s not back to the mad days of ’99 and storage vendor sales staff earning way too much money but for storage buyers, the emergence of choice without compromising on availability or simplicity has never been better.

Gavin McLaughlin