Ihave to admit that until recently, I was completely unfamiliar with the term “fail-back.” I’m still not sure if it’s even real computer lingo, but it did come up during a conversation with a client. He used it to refer to the process of resuming computer operations at the home office after a disaster recovery operation. His company was one of those affected by the blackout in the northeast. Rather than risk a prolonged outage, they decided to quickly initiate their disaster recovery plan, transferring all computer operations to their hot-site facility, not a trivial procedure.
The good news is that they were successful in restoring business operations from their remote location. There were some glitches, but for the most part, their recovery plan worked well. They were able to completely resume processing in less than 36 hours. The bad news is that their recovery planning didn’t give much thought to how operations would be brought back to the home office.
This processing disruption was relatively short in duration. Their equipment wasn’t harmed nor was their facility damaged. But when the lights went back on and they were ready to go back home, they realized that it wouldn’t be so easy. In fact, the process of returning was almost as difficult as the disaster recovery operation itself.
Transactions were processed for most of their critical applications for nearly two full days. Now files updated at this remote facility no longer matched those at the home office. To return, they were faced with either reprocessing all of these transactions or executing their disaster recovery plan in reverse; hence the term fail-back.
The situation might have been less painful if they had just given as much thought to returning from a disaster as they did planning for the disaster itself. My client is now working on his fail-back plan. His goal is to develop a way to identify, capture, and restore just the files that changed while processing at the hot site. Because only business critical applications were run at the remote facility, this solution could easily reduce the return effort to nothing more than restoring a few files.
It might be worth the time to prepare your own fail-back plan so you can go home again.
Larry Wisniewski is vice president of marketing for 21st Century Software, Inc. The company provides intelligent, enterprise-wide data protection and recovery solutions to the Global 1000 marketplace. Wisniewski can be reached at 21st Century Software, (800) 555-6845.
