Plan B is Plan A...
When it comes to disasters the wrong thing to do is to have a “recovery”. What you really want is a plan for sidestepping the unexpected disaster and moving on to Plan B without even breaking a sweat. Good planning and thinking through the potential problems before there is a problem goes a long way. Always make sure that plan B is part of the day-to-day operations or Plan A.
A few weeks ago, Amazon had an outage at one of their cloud computing data centers that is used by many companies and well known websites. Everyone hears about the outage and how the site isn’t up and everything is down, but what about those companies who considered the possibility ahead of time and stayed up? Informationweek published a great article about how some government sites were built in such a way as to seamlessly fail over to another working site. The government doesn’t disclose details on how exactly they did it, but it basically boils down to the fact that they had plan B all along and it was folded right into the day-to-day operation of Plan A (Amazon cloud).
As a professional in the cloud computing space I can totally appreciate the lengths at which providers go to in order to strive for perfect uptime despite the variable and unpredictable demands on the system while growing and scaling profitably to sustain the system. And I can also appreciate that sometimes things fail. But when they do fail, let’s make sure that we all think through the possible Plan B’s and incorporate Plan B into Plan A right from the start. And then you are taking advantage of all of the benefits of cloud computing, plus the distinct honor of being smarter than the average bear and having a site that stays up, when everyone else is down.
