Parallelization of the flow-path network model using a particle-set strategy

High-performance simulation of flow dynamics remains a major challenge in the use of physical-based, fully distributed hydrologic models. Parallel computing has been widely used to overcome efficiency limitation by partitioning a basin into sub-basins and executing calculations among multiple processors. However, existing partition-based parallelization strategies are still hampered by the dependency between inter-connected sub-basins. This study proposed a particle-set strategy to parallelize the flow-path network (FPN) model for achieving higher performance in the simulation of flow dynamics. The FPN model replaced the hydrological calculations on sub-basins with the movements of water packages along the upstream and downstream flow paths. Unlike previous partition-based task decomposition approaches, the proposed particle-set strategy decomposes the computational workload by randomly allocating runoff particles to concurrent computing processors. Simulation experiments of the flow routing process were undertaken to validate the developed particle-set FPN model. The outcomes of hourly outlet discharges were compared with field gauged records, and up to 128 computing processors were tested to explore its speedup capability in parallel computing. The experimental results showed that the proposed framework can achieve similar prediction accuracy and parallel efficiency to that of the Triangulated Irregular Network (TIN)-based Real-Time Integrated Basin Simulator (tRIBS).     

A machine learning approach for predicting computational intensity and domain decomposition in parallel geoprocessing

ABSTRACT

High performance computing is required for fast geoprocessing of geospatial big data. Using spatial domains to represent computational intensity (CIT) and domain decomposition for parallelism are prominent strategies when designing parallel geoprocessing applications. Traditional domain decomposition is limited in evaluating the computational intensity, which often results in load imbalance and poor parallel performance. From the data science perspective, machine learning from Artificial Intelligence (AI) shows promise for better CIT evaluation. This paper proposes a machine learning approach for predicting computational intensity, followed by an optimized domain decomposition, which divides the spatial domain into balanced subdivisions based on the predicted CIT to achieve better parallel performance. The approach provides a reference framework on how various machine learning methods including feature selection and model training can be used in predicting computational intensity and optimizing parallel geoprocessing against different cases. Some comparative experiments between the approach and traditional methods were performed using the two cases, DEM generation from point clouds and spatial intersection on vector data. The results not only demonstrate the advantage of the approach, but also provide hints on how traditional GIS computation can be improved by the AI machine learning.     

武汉区域气象中心数值预报业务并行计算机系统

对武汉区域气象中心并行计算机系统进行了详细地介绍，分析了并行计算机体系结构、网络和存储系统特点；给出了在并行计算机SP上实现数值预报业务并行化的部分结果；对数值预报模式在串、并行编程环境下的结果进行了分析比较。     

基于MPI的LAGFD-WAM海浪数值模式并行算法研究

在LAGFD-WAM海浪串行数值模式基础上,利用MPI信息传递机制实现其并行化。通过对模拟区域合理划分,对数据采取分块加载,实现了各个节点的负载平衡;通过对算法的改进实现了粗粒度计算,大幅度减少了通信量,从而提高了程序的执行效率。对串行计算和并行计算的效率比较表明,本文建立的方法能够得到较高的加速比。对全球海浪模式,加速比和CPU数目大致呈线性关系。对高分辨率的区域海浪模式,在128 CPU条件下加速比可以达到91.9。     

海洋石油管线探测多波束剖面声纳设计

针对海洋石油管线探测多波束剖面声纳的高速、大容量的实时信号处理任务,以及声纳系统进一步扩展的需求,设计并实现了一种基于IP网络互连的、以并行DSP为处理节点的多波束剖面声纳系统.该系统应用于海底石油管线探测与定位,系统中的每个处理节点与数据采集转换部分采用TCP/IP网络连接,可以通过物理上添加一个或多个处理节点成倍的提高系统的信号处理能力,而在软件上无须大的改动.通过电联调与水槽试验,验证了系统的可行性和稳定性,该设计为高速水声阵列信号实时处理系统提供了一种全新的解决思路.     

空间信息网格中远程空间连接查询的优化

随着空间信息网格的建设,网格平台上管理的空间信息资源越来越丰富,这促进了空间信息网格中空间数据分布式查询的应用需求,而在分布式空间查询中,空间连接查询操作往往成为性能的瓶颈.根据空间信息的特点,通过利用网格计算资源来优化空间连接查询的执行.首先基于网格服务构建网格平台分布式空间数据查询软件结构,通过设计远程空间连接执行服务利用网格平台中的计算资源;根据空间信息的特点.采用基于Kd-Tree空间分区并行连接的方法提高远程空间数据连接操作执行效率,并给出了远程空间连接执行的查询代价模型;然后根据连接代价模型设计了远程空间连接查询执行计划优化生成算法;最后总结了本文工作并探讨了下一步研究方向.     

应用改进并行遗传算法反求水文地质参数研究

利用单纯形法局部搜索速度快和遗传算法全局寻优的特点,同时为了克服两种方法各自的弊病,提出采用混合单纯形技术的并行遗传算法(Hybrid Simplex Parallel Genetic Algorithm,SPGA)进行水文地质逆问题的求解。详细阐述了SPGA算法的具体操作和实现,并将该方法应用于水源地的地下水模拟反演中。计算结果表明,SPGA算法在水文地质参数寻优计算中具有比较好的可靠性和计算效率。     

关于角平分线法建立线状缓冲区算法的改进

分析了用对角平分线法建立缓冲区的算法,提出了一种新的求角平分线和缓冲点的方法,实践证明其在算法实现方面是可行的,而且效果比较理想。     

内蒙古大青山地区寒武系与奥陶系之间的一个重要的层序界面

大青山地区寒武纪－奥陶纪地层分布零星，是由中生代逆冲推覆构造就位于此的一系列近东西向延展的构造岩片组成。下寒武统色麻沟组直接覆盖在太古界变质岩之上，缺失中元古界什那干群，寒武系与奥陶系之间呈假整合接触，标志着区内寒武纪与奥陶纪之间发生过一次抬升构造运动，其形成机制与古亚洲洋内的稳定陆快向华北板块拼贴事件有关，是早奥陶世末期华北板块整体抬升的序幕。这个层序界面的认识，对于恢复华北板块北部边缘构造演化历史具有重要意义。