Performance evaluation of networks of workstations with hardware shared memory model using execution driven simulation

Networks of workstations (NOWs) are becoming increas-ingly popular as a cost-effective alternative to parallel com-puters. Typically, these networks connect processors using irregular topologies, providing the wiring flexibility, scal-ability, and incremental expansion capability required in this environment. Similar to the evolution of parallel com-puters, NOWs are also evolving from distributed memory to shared memory programming model. However, physical distances between processors are longer in NOWs than in tightly-coupled distributed shared-memory multiprocessors (DSMs), leading to higher message latency and lower net-work bandwidth. Therefore, the network may be a bottle-neck when executing some parallel applications in a NOW supporting a shared-memory programming paradigm. In this paper we analyze whether the interconnection network is able to efficiently handle the traffic generated in a NOW with the shared memory model. In particular, we are interested in analyzing the influence of the routing mechanism in the performance of the system. We evaluate the behavior of a NOW with irregular topology by means of an execution-driven simulator using SPLASH-2 appli-cations as the input load. The results show that the rout-ing algorithm can considerably reduce the total execution time of applications. In particular, routing adaptivity can reduce the total execution time by 58 % in some applica-tions. These results confirm the behavior observed in previ-ous works using synthetic traffic loads.