Theoretical physics doesn’t often cross over with Cloud computing. However, in the world of recovery and business continuity, it’s very relevant.
Austrian physicist Erwin Schroedinger created a famous thought experiment in 1935, where he imagined a cat being placed in a sealed box alongside a radioactive element and a flask of deadly poison. The element would decay over time, but the chance of this taking place at any given moment was a probability of 50/50.
[easy-tweet tweet=”Without testing, it is impossible to tell; a case of “Schroedinger’s Backup”, if you will.” user=”mastersian and @comparethecloud”]
If the element does decay, the phial of poison would break and kill the cat. The chance that the cat was alive or dead was therefore equal too. The important lesson from this was that it was impossible to know whether Puss had survived his brush with death unless you opened the box; before this, it would be impossible to tell.
How does this apply to Cloud and disaster recovery (DR)? For many traditional implementations, recovery is just as dicey a proposition as for Schroedinger’s famous feline. Previously, testing backup and recovery programmes was hard to justify, based on the time required for systems to be down during the recovery test process. There was always the potential situation where the backup process would not actually work, leading to both lost revenue and (potentially) loss of career. Better then to leave things as they are and trust the plan, some might say.
It’s understandable that people may have this mindset. If the plan is not tested, then things as they stand can continue and everything remains operational. However, it might also fail, and it would be impossible to tell beforehand if the recovery would succeed. Without testing, it is impossible to tell; a case of “Schroedinger’s Backup”, if you will.
However, this lack of testing leads to false hope and fragile IT remaining in place. While the business may be happy in ignorance, any incident could lead to big problems in the future. The longer that operations go on, the bigger the issue will get.
Today, using Cloud can help remove the uncertainty around recovery through making testing simpler. Helping companies test their backup and recovery processes can be a great opportunity to prove the value that Cloud can deliver, as well as ensuring that all the efforts around business continuity planning are worthwhile investments.
For companies that are already using Cloud for recovery, testing the implementation can be as simple as starting up new instances and making sure that the process works effectively. The production systems should be able to carry on running while the test is carried out, as the test can be run on new Cloud infrastructure.
For those organisations that want to look at how they can test their current approach, there is some more work involved. Replicating data from the existing backup systems over to a Cloud instance would be required; this Cloud instance could then be fired up to show that the backup systems are working properly. While there might be more time required to carry out the initial replication, the result is the same.
Testing Cloud recovery programmes as standard can actually be a great differentiation point for DRaaS providers, as it shows that the provider knows exactly how their recovery procedures will work in practice. It also ensures that the Cloud service keeps up with all the changes that are taking place across IT, both within customers and in the industry as a whole.
While testing might not be high up on the agenda for the CIO, it’s an essential prerequisite for IT teams that have to be sure that their infrastructure is being protected and that recovery systems are working properly. Using the Cloud to carry out this testing makes sure that there are no excuses for not doing this.