I still remember clearly, and I’m pretty sure you do too, the Eastern Seaboard blackout of 2003. Friends of mine suffered fates as varied as being stuck in an elevator for hours, having to evacuate a subway car in darkness, and having no water for several days. I escaped relatively unscathed. My biggest complaint was that convenience store clerks couldn’t ring in my bottle of water. While, in retrospect, that certainly sounds like a First World problem, it was about 900 degrees outside.
It was an electric catastrophe of once-in-a-lifetime proportions. Except, it wasn’t. It’s happened twice in my lifetime, though I was only two when the East Coast went down in 1965 and the Bee Gees felt compelled to write The Lights Went Out in Massachusetts. My point is, catastrophes happen. And with enterprises becoming ever more reliant on technology—cashiers can’t sell you a bottle of water without the PoS system anymore —catastrophe is not an option. It’s a 24/7 business world out there, and you can’t afford your IT infrastructure going down.
The conundrum is: can you afford to ensure your IT infrastructure doesn’t go down?
Disaster recovery and business continuity are not the same thing, but they’re two sides of a coin. Business continuity is about ensuring that the enterprise can continue running, at least at a minimum level, despite a catastrophic system failure. Disaster recovery is about restoring the business to the state that existed before the failure. They’re both dependent on redundancy: parallel systems and data stores that are available should the primary systems and data stores fail.
This mirroring is technologically feasible. In fact, there are entire companies and brands built around it. For it to work in a practical business environment, there are two critical conditions to be met. First, the parallel systems must be synchronized. The second condition is that there has to be a geographic disparity. This can be as simple as residing on a different server in the same data centre, or being on different sides of the continent or even in different countries. The former is adequate to compensate for a hardware failure; the latter is necessary when the lights go out in an entire region.
Either way, it’s an expensive proposition—fully redundant hardware and software, with the accompanying overhead and maintenance costs. This is where cloud-based virtual computing environments can come to the rescue.
Virtual and cloud-based aren’t the same thing, though the latter is dependent on the former. A virtual environment packages compute, storage and I/O resources together in a simulation of a physical machine. Cloud computing frees that virtual machine from the confines of a single piece of hardware, able to move from server to server as necessary.
This environment is ideal for disaster recovery. It’s ideal for business continuity too—a duplicate of your business processes is available immediately by simply spinning up a copy of System 1. And, you’re shifting costs from capital expense (CAPEX) to operating expense (OPEX), which delivers some cost certainty.
Disaster recovery and business continuity aren’t an option in a 24/7 business world, but there are at least cost-effective options to building duplicate hardware and software stacks.