As parallel systems scale up, the energy consumption of high performance computing systems drastically in-creases. The use of high power and energy leads to many problems, such as low reliability, bad stability and so on. In this case, the energy consumption problem draws unprecedented attention. Thus it is necessary to take into account the energy consumption while pursuing high computing performance in system design and research. This brings a new challenge for building a high performance computing system metric model. The measure index for the performance of the large-scale sys-tem demands a shift from "high performance" to "high productivity". Then, this paper uses speedup, synthetically meas-ures the computing performance and energy consumption in view of system scalability, and proposes a power parallel speedup for evaluating the comprehensive performance of high-performance computing systems. The model can measure the pro-ductivity of parallel systems and guide system design and application research. Finally, we validate the model of power par-allel speedup through analysis and simulation.%随着并行系统规模的扩大,高性能计算系统运行时消耗的能耗也在急剧增长,过高的能耗也给系统的可靠性、稳定性等方面带来严峻挑战.在这种情形下,能耗问题受到了前所未有的关注.因此,设计和研究高性能计算系统,需要在考虑高计算性能的同时兼顾系统低能耗的要求,这为高性能计算系统的度量模型提出了新的挑战.于是,大规模并行系统逐渐从"高性能"走向"高效能"的衡量标准.基于此,本文采用加速比度量指标,从系统可扩展角度将计算性能和能量消耗要素进行综合,提出了一种度量高性能计算系统综合性能的能耗并行加速比模型.该模型能够直观地反映并行计算系统的效能,旨在指导系统设计和应用研究.最后,通过对该模型的分析和模拟,验证了模型的有效性.
展开▼
