Execution Based Evaluation of Multistage Interconnection Networks for Cache-Coherent Multiprocessors

In this paper, performance of multistage interconnection network with wormhole routing and packet switching has been evaluated for cache-coherent shared-memory multiprocessors. The traffic in cache-coherent systems is characterized by traffic bursts, one-to-many and manyto -one traffic, and small fixed length messages. The evaluation is based on execution-driven simulation using various applications. The comparison of execution time between a packetswitched and a wormhole-routed network shows that the wormhole-routed network provides significant advantage. We also evaluate wormhole networks for variable number of virtual channels and flit buffers per channel. 1 Introduction Large-scale shared-memory multiprocessors are difficult to design but they provide a unified view of the memory for easy programming. These systems are built using processor-memory nodes that are connected through an interconnection network (IN), such as 2-D torus, hypercube or multistage interconnection network(M..