An analytical model to evaluate the performance of cluster architectures
Léonardo Brenner, César A. F. De Rose, Paulo Fernandes · 2002
This paper presents a rather simple method to model cluster architectures using Stochastic Automata Networks (SAN) formalism. Although, the modeling power of SAN, the presented models are quite simple and do not fully describe the complexity of existing clusters. It is not the purpose of this paper to proposed ready-to-evaluate models to real clusters, but just introduce the formalism to the parallel architectures community. The conclusion of this paper discusses the further steps in order to perform a real case evaluation. 1.