The authors examine the various kinds of distributed systems and discuss some of the reliability issues involved. They first concentrate on the causes of unreliability, illustrating these with some general solutions and examples. Among the issues treated are interprocess communication, machine crashes, server redundancy, and data integrity. Then they examine one distributed operating system, Amoeba, to see how reliability issues have been handled in at least one real system, and how the pieces fit together.
|Title of host publication
|Proceedings - 6th Symposium on Reliability in Distributed Software and Database Systems
|Number of pages
|Published - 1987