It’s that season again! For those who don’t know, I live in Florida, and yes, we do have seasons down here. We have Tourist Season, Lovebug Season, and of course the one we’re best known for: Hurricane Season. With Hurricane Season starting on June first, most companies should have finished their preseason disaster recovery test by now. Oh, how things have changed from the days of flying off to have loads of fun at the remote data center, restoring test objectives from tape to prove we had the capabilities to restore applications and services slated for that specific test.
One of my greatest “oh yeah, now we’re talking” moments was the first disaster recovery test we had after introducing virtualization to production. We had SAN-based replication of the virtual machines, and I pretty much had most of the week as a vacation in Philadelphia for that test. I had just about all my tasks scripted. As soon as the hosts were built and storage presented, I kicked off a script that registered and started all the virtual machines while I went down the street for a classic Philly cheesesteak. What a great week! As for my peers working their physical machine recovery—well, let’s just say their week was not as much fun, to say the least. I did have to do my recovery three times that week, but no worries: I had plenty of time to spare.
Now, that was pretty old school. In today’s modern hybrid world, what used to be in-house applications—like email, content management, file shares, and in some cases much more—have evolved from running in-house to running in multiple offsite clouds. This is naturally going to change the way we look at and approach disaster recovery in today’s world, and this approach will continue to evolve as cloud technology matures.
There is an assumption that once you migrate your application, service, or both away from your corporate data center, backups and recovery options are handled by the third party—but I do hope you verify that before you make the move. For the most part, you would be correct in that even if your corporate office were made unavailable due to some kind of natural disaster, the services should continue uninterrupted. But what if something happens to the application? Again, in most cases you should be able to restore to an earlier point in time, and depending on the budget, other options could be available.
This brings me to the point I want to focus on. With the adoption of hybrid clouds in our environment, have we completely lost control of our data? Do companies still have the capabilities to centrally restore the enterprise if needed? I believe that in most cases, the answer to that question is “no.”
In today’s world and looking to the future, do we even need to worry about that? I would like to think that we still have the option and the ability to access all the data available for all kinds of recovery, whether from some kind of disaster or something completely different. Companies have given up control of the data and, in some ways, control of their infrastructure. So, let me remind you one more time: it’s that season again!