Summary

Building a highly available and fault-tolerant system requires you to carefully evaluate both your requirements and your resources to eliminate single points of failure within the system. You should evaluate each of the hardware subsystems within the overall system for fault tolerance, and ensure that recovery procedures are clearly understood and practiced to reduce recovery time in the event of a failure. Uninterruptible power supplies, RAID systems, distributed file systems, and clustering are all methods for improving fault tolerance. In the next chapter, we discuss the registry: what it is, how it's structured, and how to back it up and restore it.