SKY's Health Maintenance Strategy Yields High Application Availability

There are two main methods that can be employed to build a system that provides High Application Availability (HAA). The first is to construct components, both hardware and software, that exhibit a very long mean time between failures (MTBF). The longer this time, the better the system is at supporting HAA. Systems that exhibit high MTBF are very expensive to build, so much so, that their cost can be prohibitive. For example, one can build a system using triple redundancy. If the system's baseline processing requirements called for several hundred processors, tripling the number of processors in order to provide a high MTBF would probably make the system too costly.