Exascale computing is one of the major challenges of this decade,and several studies have shown that communications are becoming one of the bottlenecks for scaling parallel applications.The analysis on the characteris...Exascale computing is one of the major challenges of this decade,and several studies have shown that communications are becoming one of the bottlenecks for scaling parallel applications.The analysis on the characteristics of communications can effectively aid to improve the performance of scientific applications.In this paper,we focus on the statistical regularity in time-dimension communication characteristics for representative scientific applications on supercomputer systems,and then prove that the distribution of communication-event intervals has a power-law decay,which is common in scientific interests and human activities.We verify the distribution of communication-event intervals has really a power-law decay on the Tianhe-2 supercomputer,and also on the other six parallel systems with three different network topologies and two routing policies.In order to do a quantitative study on the power-law distribution,we exploit two groups of statistics:bursty vs.memory and periodicity vs.dispersion.Our results indicate that the communication events show a“strong-bursty and weak-memory”characteristic and the communication event intervals show the periodicity and the dispersion.Finally,our research provides an insight into the relationship between communication optimizations and time-dimension communication characteristics.展开更多
基金funding from the National Key Research and Development Program of China(2017YFB0202200)the Advanced Research Project of China(31511010203)+1 种基金Open Fund(201503-02)from State Key Laboratory of High Performance Computing,and Research Program of NUDT(ZK18-03-10).
文摘Exascale computing is one of the major challenges of this decade,and several studies have shown that communications are becoming one of the bottlenecks for scaling parallel applications.The analysis on the characteristics of communications can effectively aid to improve the performance of scientific applications.In this paper,we focus on the statistical regularity in time-dimension communication characteristics for representative scientific applications on supercomputer systems,and then prove that the distribution of communication-event intervals has a power-law decay,which is common in scientific interests and human activities.We verify the distribution of communication-event intervals has really a power-law decay on the Tianhe-2 supercomputer,and also on the other six parallel systems with three different network topologies and two routing policies.In order to do a quantitative study on the power-law distribution,we exploit two groups of statistics:bursty vs.memory and periodicity vs.dispersion.Our results indicate that the communication events show a“strong-bursty and weak-memory”characteristic and the communication event intervals show the periodicity and the dispersion.Finally,our research provides an insight into the relationship between communication optimizations and time-dimension communication characteristics.
