I’m actually old enough to remember when people looked forward to airline travel. Yes, as hard as it is to believe, there was a day when you could check in at the counter, get your ticket and walk, unmolested, to your departure gate, stop and purchase a tuna fish sandwich for $10.50, and then, just a few hours later, be greeted by a gaggle of your loved ones as you walked off the plane.
Now, of course, things are a little different. You can still pick up your ticket at the counter, although the smartphone boarding pass is easier. But that is just a prelude to enduring a wait only slightly shorter than that of an incoming refugee behind a person who hasn’t flown since the Wright Brothers landed at Kitty Hawk (“I have to take my shoes off?”), then potentially receive a pat down that is only mildly less invasive than your last proctology exam, and proceed to your gate while preparing your strategy to stake your claim to a portion of the overhead baggage compartment before that grandmother from Cleveland cleans and jerks her 100-pound wheely bag into the last available spot before you.
Of course, no one is there to meet you at the gate, since if anyone tried, they’d be beat down by a plethora of polyester clad TSA workers whose lack of enthusiasm is only eclipsed by their desire to publically humiliate any and all violators of security line etiquette: “I said, all laptops need to be placed in a separate bin.” All of that is bad, but what’s worse is that this modern day version of the Bataan Death March is all too often leading to the dreaded cancelled flight.
Of course, there has always been the occasional canceled flight, but historically this was due to the more extreme “acts of God.” But increasingly it is outages and “glitches” within the data center that are responsible for these large contingents of stranded travelers. For example, recently 250 sets of passengers who had entrusted Delta to get them from point A to point B were profoundly disappointed when an outage led to their flights cancellation. This event occurred only a week after a similar misfortunate “issue” forced a number of United passengers to make alternative travel arrangements. And Delta suffered a previous disruption that wreaked havoc on travel schedules this past August, as well. Each of these incidents could be traced back to the data centers.
Airline travel aside, it is indisputable that more and more of our lives are governed by the operations that take place on millions of square feet of computing floors across the country and the globe. With so much of our present — and certainly our future — dependent on the uninterrupted operation of public and private data centers, the ability to build more reliable architectures and the ability to keep them secure will become, perhaps, the two key drivers of data center operational planning and network construction.
While these elements have always been crucial to successful data center performance, the margin for error continues to diminish with each new application included within their walls. While the integrity of airline schedules is a non-trivial matter, it is difficult to imagine the economic impact of similar outages and issues arising in facilities supporting IoT applications for example. Perhaps paradoxically, our dependence also increases our vulnerability. No one wants to contemplate the time, sometime in the near future, when a data center issue renders your refrigerator incapable of calling your phone to tell you you’re out of Heineken, for example.
Our relationship with technology has always been a fickle one where we quickly become dependent on the next new development, so the intrinsic nature of data centers, and the applications they support, in our lives is not surprising. This co-dependent relationship continues to positively impact every aspect of our lives. It also dramatically escalates the requirement for fault-proof operation, since as our dependence grows, the impact of a service interruption has the potential to be much more irritating than a canceled flight.