Planning for Disaster Recovery
We don’t like to think about disasters, but preparing for them is better than ignoring them. FEMA estimates that 40% of businesses that experience a disaster never open again, and 25% last no more than a year after that. One of the big reasons for this is data loss. If a fire or hurricane destroys a business’s data records beyond recovery, the chances of putting the pieces back together are poor. However, a business with a disaster recovery plan can make it through a catastrophe with its data intact and get up and running quickly.
The key elements for disaster recovery are data backup, restoration to new systems, and re-establishment of services. Each of these requires consideration on its own. Simply having a backup doesn’t guarantee being able to become operational again without problems.
The first step is to identify all critical systems. A disaster recovery plan needs to specify what data systems are essential to operation and how they fit together. This includes listing the machines, describing their hardware, and characterizing the network that they participate in. If any machines are so old that it would be hard to replace them, the recovery plan needs to spell out a way to find replacements or migrate quickly to an alternative. For instance, if SyQuest disk cartridges from the 1990s are essential to the system’s operation, the recovery plan needs to note this and indicate whether it’s possible to find replacement hardware or devise an alternative.
It’s necessary to back up the data on the critical systems to storage far enough from the premises that the backup will survive. The backup needs to include a system image of any bootable disks, allowing loading of the entire file system onto a new machine that can run it.
Planning the restoration process is especially difficult. Affected hardware will need replacing. The restoration will need to load the saved system images onto new computers or drives. The network needs to incorporate the replacement devices so that they function the way they did before. It’s like putting together a jigsaw puzzle that’s fallen apart, where some of the pieces are missing and you have to draw them over again.
People are central to a disaster recovery plan, and how well they coordinate their actions will determine whether bringing up the data systems is a smooth process or a confused one. Someone needs to be in charge of recovery, and a backup person should be available in case of necessity.
Not everyone may be reachable if a disaster occurs. A way to contact everyone outside the usual channels is necessary, and the plan shouldn’t fall apart if any one person is incapacitated or out of reach. The information for restoring the systems needs to be in a safe place, not one that a disaster will destroy along with the rest of the office. The password for the backup site should be in a locked box in a place that’s protected but away from the main facilities.
A recovery plan needs testing. Doing a full simulated recovery is difficult, but testing as much of the process as possible will improve confidence in it and help to discover any omissions. All the people involved in the plan should know what they’ll have to do if a disaster strikes. Speed of recovery is important, and every day that passes means lost revenue and customer confidence. A tested plan will flow more smoothly and lead to a quicker recovery.
The plan needs periodic review, at least once a year. The configuration of essential machines will change. Some devices may have grown obsolescent, and the plan for replacing them may need updating. Business requirements change over time. An up-to-date plan will take all recent changes into account.
Contact BWS Technologies to learn more about the support that we offer in everything from regular maintenance to disaster planning.