An algorithm for the Classification of Coherence Related Overhead in Shared-Bus Shared-Memory Multiprocessors

This paper describes an algorithm for the classification of coherence overhead in shared-bus shared-memory multiprocessors. This algorithm is applicable to Write Update, Write Invalidate and Hybrid protocols and models the effect of finite size (real) cache. It differs from previous classifications because it classifies any sources of coherence overhead, i.e. invalidation miss, invalidate and write transactions, and also classifies the sharing due to process migration, namely passive sharing. A validation has been carried out considering SPLASH benchmark, and sample results in the case of E-Commerce Server are given.