A comprehensive modeling for performance evaluation of regular interconnection network

Interconnection network is one of the major factors which determine the performance of a multiprocessor system. Therefore, it is imperative to have a model by which the efficiency of the interconnection network can be correctly evaluated for various application environments. The authors present an interconnection network evaluation mode where the memory bandwidth and processor acceptance probability are considered primary measures. The proposed analytical model includes typical network control policies such as bus arbitration and rejected request handling policies as well as home memory concept. The crossbar network is selected to test the model and computer simulation validates the correctness. The study also confirms that the home memory and dynamic bus arbitration policy improve the network performance