A distributed computing system is one consisting of several computers operating in tandem, possibly forming one logical unit from the perspective of the users. In a more precise sense, we can say that a distributed system is a set of processors connected by some form of communication medium that allows them to work on different parts of the same problem, or to perform multiple tasks at the same time. The computing resources that form a distributed system may be geographically widespread, and may have to operate with no central authority or guarantees of performance.

In this situation, it is not difficult to see why it would be important for faults in systems to be tolerated. Part of the interest is motivated by questions of safety in distributed systems--what constitutes a safe distributed system, and how can we create distributed systems that are safe? Tasks and services...