Like many IT endeavors that have slowly evolved over decades, so too has the way we store, access, protect and utilize file data within our organizations. Like all things that evolve in this manner, we find ourselves doing things in ways no one would ever do if they were starting new.
In the following video brief, ESG Founder and Senior Analyst, Steve Duplessie discusses the current state of enterprise file storage and why legacy systems are no longer effective at handling today’s demanding file workloads.
Like many IT endeavors that have slowly evolved over decades, so too has the way we store, access, protect, and utilize file data within our organizations. And like all things that evolve in this manner, we find ourselves doing things in ways no one would ever do, if they were starting new. Let’s start with a concept that I’m sure you’ve heard, “I loved my first filer…but I hated my 30th.” Sure it’s funny but, sadly, it’s true. And here’s why. What hasn’t evolved or changed substantially, over the last few decades, is what we wanted to happen, only how it happened. What we or any sane person really wanted out of their file environment is fairly easy to define. We want our files to be centrally stored, easily located, perform as fast as we need them to be, to be protected, and to allow us to access our files from wherever we happen to be. Seems simple enough.
We solved this problem by building a dedicated, specialized box. NetApp became a huge company, by developing this box such that IT shops could consolidate NFS data on one system and not have to have zillions of files scattered all over dozens or hundreds of workstations. It made managing the file infrastructure much simpler, easier to protect and it could be optimized for performance. It’s exactly why people loved their first system. The problem is that data growth eventually forced us to buy another system. And two is harder to deal with than one. Once we had more systems, we needed to buy even more, just to protect the ones we already had. Or since our employees don’t all work in one office, we needed to distribute more and more systems to where the users were, for performance reasons, which made everything else harder to manage, protect, and to access. Our desire or what we wanted to have happen in our file world hasn’t changed since the beginning of time. The way we accomplish those goals is what’s changed.
In the old days, meaning like last year, we simply couldn’t do what we actually wanted. We were forced to do things poorly or not do them at all. But that isn’t the case anymore. Today, if you’re starting a company and you are the 17‑year-old CEO, your demands for your file infrastructure would be the same as the stodgy 53‑year-old CEO 30 years ago. Namely, “I want my files centralized, ideally on one single system. And that system can grow as big as it has to grow, forever.” You want to manage one single thing and one thing only. You want your files to be accessed from anywhere on the planet and perform as if the server is right underneath your desk. You want your files to always be available. You want to be able to seamlessly collaborate, which is a fancy term for describing that you want the file system you see in Boston to be the exact same file system that your colleague sees and accesses in Sydney, Australia. And, when you open up shop in Barcelona, you access the exact same file system, at the exact same performance levels. You want the system to self-protect. You don’t have any interest or desire to replicate data or systems, just to protect them. You want perfect economics. You never want to overbuy or prepay for things that you aren’t using. You want to pay for exactly what you use and want to take advantage of the never-ending cost reductions available for capacity, compute, and access. You don’t want to have to wait years to refresh. You want it all now.
If you’ve been working in the file world of IT for more than a decade, you’re probably snickering right now. But if you let your mind wander, you realize these are not crazy demands created by shortsighted millennials. This is exactly what you would demand. Think of it this way. If NetApp went to its customer base and said, “For a million dollars in cash, I’ll take everything you have out, put one über-NetApp-system in place that will scale in any dimension, forever, charge you for what you use only, put it in one data center but it will service every single user you have, no matter where in the globe they are, and they will get local performance, inside a single global file system, which is completely self-protecting, so you don’t need to deal with backup and recovery, both from a cost or an operation perspective.” 99% of those customers would sign up on the spot, because that’s what every sane person would want.
And this is exactly what Nasuni has built. It’s not a filer. It’s not even a box. It’s an intelligent file infrastructure architecture, designed for what the world actually needs today. I’m not disparaging NetApp. I love them. They revolutionized how files are dealt with inside of IT. But they are now creating the same problem they originally solved, through no fault of their own. Data growth is relentless. If we had no data growth, we would have none of these problems. We’d build one box and be done with it. But since never-ending data growth is the only true constant we face, adding box after box has been the only way, to date, at least, to solve these problems. But if you didn’t have to do it that way, you most likely wouldn’t. Nasuni was lucky enough not to have a million customers and billions in sales to protect, if you can call that lucky. They had a clean slate, and, as such, could address the modern problems with modern solutions. And that’s what I find most interesting. If all these other problems really go away, what more could we accomplish in these environments? Thanks.