R. Smeikal and K.M. Goeschka (Austria)
fault-tolerance, distributed systems, availability, consis tency, replication
The first contribution of this paper is to clarify a confusing mixup of three related terms throughout the literature in the field of distributed object-based fault-tolerant comput ing: Replica consistency, which defines the correctness of replicated data, concurrency consistency, which defines the correctness of concurrent access to a single set of replicas, and constraint consistency, which defines the correctness of the system state with respect to the set of constraint condi tions. Based on these terms, the second contribution of this paper is to examine a very specific aspect of fault-tolerant distributed systems: the explicit trade-off between avail ability of replicated objects and constraint consistency. If the system faces site crashes or network partitions, less but well controlled constraint consistency is accepted to gain higher availability of objects.
Important Links:
Go Back
