This volume presents papers given at a workshop on fault tolerance. The contributors discuss the general experiences of industrial practitioners, provide a profile of the current state of research, and identify areas of continuing importance in the years to come.

Top five challenges facing the practice of fault-tolerance.- Fault tolerance cost effectiveness.- Fault tolerant applications systems; A requirements perspective.- Scalable shared memory multiprocessors: Some ideas to make them reliable.- Application of compiler-assisted rollback recovery to speculative execution repair.- Fault tolerance: Why should I pay for it?.- Stable disk — A fault-tolerant cached RAID subsystem.- Simple design makes reliable computers.- Roll-forward checkpointing schemes.- Fault-tolerant architectures — Past, present and (?) future.- A highly available application in the transis environment.- Reliable enterprise computing systems.- Fault tolerance for clusters of workstations.- Two techniques for transient software error recovery.- Software-faults: The remaining problem in fault tolerant systems?.- Fault tolerance enablers in the CHORUS microkernel.- A reliable client-server model on top of a micro-kernel.- Distributed fault tolerance — Lessons learnt from Delta-4.- Arjuna and Voltan: Case studies in building fault-tolerant distributed systems using standard components.- Fault tolerant platforms for emerging telecommunications markets.- Fault-tolerance in embedded real-time systems.- The systematic design of large real-time systems or interface simplicity.- Fault tolerance in embedded real-time systems: Importance and treatment of common mode failures.- Highly-available data services for UNIX client-server networks: Why fault-tolerant hardware isn't the answer.- The management of replicated data.