基于MPI+OpenMP的水声传播并行预报方法

Parallel Forecast Method of Underwater Acoustic Transmission Based on MPI+OpenMP

基于SMP集群,利用MPI+OpenMP混合并行编程方法,设计和开发了水声传播产品多级混合并行预报算法,该算法结合SMP集群的体系架构特点,通过节点内共享内存,节点间消息传递的方式,充分发挥两种模式各自的优势,克服MPI模式进程间通信时间长、OpenMP模式扩展性差的缺点,并对程序进行了优化处理,实现了水声传播产品的快速高效预报。测试结果表明,该方法可充分发挥两种并行编程模式的优点,有效提高水声传播产品预报程序的并行效率和扩展性,大幅降低预报时间。

In order to complete the calculation of underwater acoustic propagation quickly and accurately,and to meet the needs of the design and use of underwater acoustic sensors for large-scale and long-time series of underwater acoustic propagation information.Based on SMP cluster and using MPI+OpenMP hybrid parallel programming method,a multi-stage hybrid parallel prediction algorithm for underwater sound propagation products is designed and developed.Combined with the architecture characteristics of SMP cluster,the algorithm gives full play to the respective advantages of the two modes by sharing memory within nodes and message transmission between nodes,and effectively overcomes the shortcomings of long inter-process communication time of MPI mode and poor scalability of OpenMP mode.The program is optimized to realize the fast and efficient prediction of underwater sound propagation products.The test results show that this method can give full play to the advantages of the two parallel programming modes,effectively improve the parallel efficiency and expansibility of the underwater acoustic propagation product prediction program,and greatly reduce the prediction time.