为提高仿真系统的运行效率,结合多核计算环境和基于反射内存网计算环境的优势,研究基于共享内存的高性能RTI(High Performance Run Time Infrastructure Based on Shared Memory,HP-RTI-SHM)系统是当前的一个热点。然而现有相关研究主...为提高仿真系统的运行效率,结合多核计算环境和基于反射内存网计算环境的优势,研究基于共享内存的高性能RTI(High Performance Run Time Infrastructure Based on Shared Memory,HP-RTI-SHM)系统是当前的一个热点。然而现有相关研究主要集中在优化通信机制等方面,其它方面还有待深入研究。在开源项目SHM-CERTI基础上,结合基于共享内存并行计算环境的特点,主要工作包括:(1)优化的软件系统结构;(2)联邦成员时戳下限计算并行优化方法;(3)联邦成员间数据交互并行优化方法;(4)基于实时操作系统的仿真任务独占式调度方法;(5)初步实现了COSIM-RTI-SHM原型系统。与SHM-CERTI的性能对比测试表明COSIM-RTI-SHM提高了仿真推进速度和仿真数据交互的速度和吞吐量。展开更多
The examination work clarifies a practical, superior registering stage for the parallel usage of the FDTD calculation on PC bunches utilizing the message-passing interface (MPI) library, which is a neighborhood framew...The examination work clarifies a practical, superior registering stage for the parallel usage of the FDTD calculation on PC bunches utilizing the message-passing interface (MPI) library, which is a neighborhood framework comprising of various interconnected (PCs), and is now generally utilized for parallel figuring. In this paper, we describe the essential elements of a parallel algorithm for the FDTD method using the MPI (message passing interface). At present, the Internet of Things (IoT) has attracted more and more researchers’ attention. Parallel FDTD method is applied to analyze the electromagnetic problems of the electrically large targets. This paper presents the concept of “the optimum virtual topology” for MPI based parallel FDTD. Parallel FDTD method is applied to analyze the electromagnetic problems of the electrically large targets on super computer.展开更多
文摘为提高仿真系统的运行效率,结合多核计算环境和基于反射内存网计算环境的优势,研究基于共享内存的高性能RTI(High Performance Run Time Infrastructure Based on Shared Memory,HP-RTI-SHM)系统是当前的一个热点。然而现有相关研究主要集中在优化通信机制等方面,其它方面还有待深入研究。在开源项目SHM-CERTI基础上,结合基于共享内存并行计算环境的特点,主要工作包括:(1)优化的软件系统结构;(2)联邦成员时戳下限计算并行优化方法;(3)联邦成员间数据交互并行优化方法;(4)基于实时操作系统的仿真任务独占式调度方法;(5)初步实现了COSIM-RTI-SHM原型系统。与SHM-CERTI的性能对比测试表明COSIM-RTI-SHM提高了仿真推进速度和仿真数据交互的速度和吞吐量。
文摘The examination work clarifies a practical, superior registering stage for the parallel usage of the FDTD calculation on PC bunches utilizing the message-passing interface (MPI) library, which is a neighborhood framework comprising of various interconnected (PCs), and is now generally utilized for parallel figuring. In this paper, we describe the essential elements of a parallel algorithm for the FDTD method using the MPI (message passing interface). At present, the Internet of Things (IoT) has attracted more and more researchers’ attention. Parallel FDTD method is applied to analyze the electromagnetic problems of the electrically large targets. This paper presents the concept of “the optimum virtual topology” for MPI based parallel FDTD. Parallel FDTD method is applied to analyze the electromagnetic problems of the electrically large targets on super computer.
