Coming or Going?
My wife likes to read People Magazine, and I’ll confess I’m not immune to sneaking a peak every now and then—who doesn’t want to know why Ben and Jen finally decided to go through with the divorce? Over time, I’ve noticed a pattern in the magazine’s editorial content. Normally, the big feature teased on the magazine’s cover is about the life altering event that some “star” has just experienced or overcome. Although they never address the really big issues, like how does Kim Kardashian make it through the average day with an IQ that doesn’t reach room temperature, there’s usually enough turmoil in the lives of the rich and famous to keep things fresh and topical. Periodically, however, there’s a lull in the action and they need to offer up a topic that they know their readers are always interested in, which, in this case, is weight loss stories. Nothing happening in Tinseltown, let’s put a few regular folks on the cover with a headline like, “I lost 200 pounds and now my life is great” to keep our readers happy until we can publish the next Brad and Angelina update. We have the same type of cycle in the data center industry—when in doubt, talk about the future of the enterprise data center.
If you’ve been paying attention lately, we’re in a bit of a slow news cycle: green, someone’s building an enormous data center, big acquistion, Open Compute Project, wash, rinse and repeat, so, obviously, it’s time to call in the old standby. Fortunately, there continues to be conflicting opinions on the subject which always makes for some good copy. As usual, the everything is going to the cloud guys weighed in with their regular position: enterprise data centers will go the way of the dinosaur because, you guessed it, everything is going to the cloud. Points for consistency, guys. The counter offered to this all or nothing position is that, yes, most everything will go to the cloud, but companies will still want to keep some things in their own facilities because some stuff shouldn’t be in the cloud. For those looking for a more fact-based perspective, the good people at the Uptime Institute have released the results of their annual survey and they offer a clearer picture of what’s going in between enterprise data centers and the cloud.
The key findings of UI’s survey of 1000 IT professionals worldwide indicate that the expiration date on enterprise facilities is probably longer than what we’ve been led to believe. For example, the respondents indicated that nearly 66% of their assets are still deployed in their own data centers, 22% in multi-tenant locations and 13% in the cloud. I think the big takeaway here is that while migration to the cloud will continue unabated, ultimately making enterprise data centers the equivalent of something found in the fossilized remains from the Pleistocene era is taking a little longer than expected. Of course, the Pleistocene lasted about 2 million years so, based on that scale, no one can say that cloud migration is proceeding at a glacial pace. The survey also indicated that most data center budgets in 2017 would either increase or stay the same, a finding that lends further credence to the folks on the coexistence side of the argument. This moderate position is further bolstered by the overwhelming rejection of the concept of deploying lower redundancy data centers. “We’ve got important stuff in here, and we can’t afford for it to go down” or “we figured out that saying software level redundancy is easier than actually doing it.”
While the UI’s survey results do a good job of inserting some objective perspective into the future of enterprise data centers, I think what they tell us is that no, the cloud will never entirely replace enterprise facilities, but the real question from the coexistence perspective is, to what degree will they? The good news for all of us is that answer to this question is still quite away off, thereby guaranteeing us that we can all look forward to a future punctuated by the data center equivalent of 200-pound weight loss stories.