THE BOUNDARY-RESTRICTED COHERENCE PROTOCOL FOR SCALABLE AND HIGHLY AVAILABLE DISTRIBUTED SHARED-MEMORY SYSTEMS

Larger size networks require Distributed Shared Memory (DSM) coherence protocols which scale well, Fault-tolerance in terms of high availabi lity is required for data access and for uninterrupted DSM service sin ce large-scale environments have a greater number of potentially malfu nctioning components, We present a new class of coherence protocols fo r DSM systems whose instances offer highly available access to shared data at low operation costs. The protocols proposed scale well; an inc rease in the number of client sites does not increase the operation co sts after a certain threshold has been reached, The results presented in this paper give strong guidelines for the overall design of DSM sys tems which offer highly available, uninterrupted services.