CertiGuide to Network+  9  Chapter 1000:  Security in the Real World       9  II  Minimizing Risk

II  Minimizing Risk 1 Disaster Recovery

Sometimes machines fail. As you probably know, they tend to pick quite inconvenient times to do so (where did you absolutely have to go, the last time your car refused to start?). One facet of risk minimization is to make sure that these failures have as small as possible an effect on business. Recall that in the chapter on Network Operating Systems, we discussed a storage technology known as RAID, and a clustering technology known as System Area Networks. One of the benefits of both of these technologies is that they provide fault tolerance, which involves a system being able to respond gracefully to failure (ie, to be tolerant of faults).

Fault tolerance is often accomplished by redundancy, as in disk mirroring, multiple-controller setups (disk duplexing) and providing “hot spares” of critical parts or entire systems. It can also be accomplished by mechanisms that can recover from errors that occur, such as RAID-5’s ability to recover data from one of its disks, when a disk fails, due to error-correction information stored across the other drives. Part of fault tolerance also involves keeping regular backups of information stored on disk, so that even if an entire computer goes belly-up, you can move the information to a new system by restoring a recent backup, and resume work. Implementing clustering is another way to increase a server’s fault tolerance. In clustering, two systems are connected to the same peripherals (such as disk drives), and run special software that allows each to take over for the other in the event of failure.

Another area of fault tolerance to consider is what happens to your IT operation if your area/building is without power for hours, or even days, at a time? Many office buildings provide backup generators to assist businesses with maintaining critical functionality even in the event of a power failure. If your office building doesn’t provide a backup generator, and it is business-critical for your systems to be up even if the power isn’t on, strongly consider investing in your own generator, or making an agreement with a local supplier of electrical generators, to provide one to you on short notice should it ever be required.

II  Minimizing Risk 1 Disaster Recovery