IT disaster recovery is about saving IT, but it’s also good to see how IT itself can feed back into IT DR, offering new concepts and ideas for moving forward.
DevOps is one such concept. You may have met it as the collaboration of IT development and IT operations, helping the two traditionally separate departments work better together to speed viable system releases, automate configuration management and deployment, and set up virtuous feedback loops.
These elements all have their place in DR, but there are other aspects of DevOps that carry through as well.
Looking at DevOps as a fusion of principles from Agile, Lean, and automation, we can pick out three main notions. First, there is flow.
This is understanding how you fit into a value chain that stretches from start to finish, to loop back again. IT disaster recovery is part of a value chain too, getting systems running again, communicating properly between themselves, and enabling users to get back to productive work.
Second, there is feedback. Fast feedback loops contribute to spotting and eradicating poor quality as rapidly as possible and in minimising the wastage of effort.
That’s the version for DevOps, but it applies to disaster recovery also. DR teams cannot work in a vacuum. They need to hear if users have been able to return to work and if they have recovered all the IT facilities they had before.
Third comes learning. In DevOps, hypotheses and predictions are made, then tested. The results help DevOps teams to learn and improve.
The same approach works for disaster recovery too. Table-top exercises and simulations show how procedures should work and highlight gaps or flaws that can be fixed before a real disaster happens.
DR teams would probably have much to gain from looking at other aspects of IT and other areas of the business too.
After all, why reinvent the wheel if there are good models already in existence to be followed?