Evaluation of various node configurations for fine-grain multithreading on stock processors

Jin-Soo Kim, Soonhoi Ha, Chu Shik Jhon
Proceedings High Performance Computing on the Information Superhighway. HPC Asia '97  
It becomes more and more interesting to construct multithreaded parallel machines using stock processors due to their high pe@ormance/price ratio. Howevel; no quantitative analysis has been reported on the effectiveness of various node conjgurations and its impact on the overall performance. In this papel; we explore three different node conjigurations in detail and compare their dynamic characteristics through the instruction-level simulation with six benchmark programs. Our experiments show
more » ... at employing a dedicated processor for communication and synchronization is a reasonable approach because it can almost double the performance. Several factors that limit the overall speedup are also presented.
doi:10.1109/hpc.1997.592172 fatcat:sjkvklxiwzhptcvqxnok2lu2ba