A load-balancing strategy for data domain decomposition in parallel programming libraries of raster-based geocomputation
A load-balancing strategy for data domain decomposition in parallel programming libraries of raster-based geocomputation
复制标题
栅格地理计算并行编程库中数据域分解的负载均衡策略
DOI：
10.1080/13658816.2021.2004603
复制
发表时间：
2021-11
影响因子：
5.7
通讯作者：
Zhu A-Xing
中科院分区：
文献类型：
--
作者：
Wang Yu-Jing;Ai Bei-Bei;Qin Cheng-Zhi;Zhu A-Xing
摘要并行编程库通过为用户隐藏并行编程细节来简化基于栅格的并行地理计算的编程。然而，由于地理计算固有的空间数据分布不规则、计算量空间变异等特点，现有库采用的数据域分解策略往往导致负载不平衡，从而影响其并行性能。基于空间计算域的概念，提出了一种栅格地理计算并行程序库中数据域分解的负载均衡策略。空间计算域的概念是基于地理计算特性的计算强度分布特征。通过使用消息传递接口（MPI）实现所提出的策略，升级了一组跨不同并行计算平台（称为PaRGO V2）的并行基于栅格的地理计算算子，以提高负载平衡并行化。该策略通过并行化两个典型的地理计算算法（即反距离权重插值和模糊c-均值聚类）使用PaRGO V2不均匀分布的计算强度进行评估。结果表明，与以前采用的数据域分解策略相比，PaRGO V2的策略在负载平衡方面有了显著的改善（即更好的并行性能）。
ABSTRACT Parallel programming libraries have been proposed to simplify programming for parallel raster-based geocomputation through hiding parallel programming details for users. However, the strategy of data domain decomposition used in existing libraries often leads to load imbalance owing to inherent characteristics of geocomputation including not only irregular spatial data distribution, but also spatial variation in the amount of computation, thereby impeding their parallel performances. This paper thus proposes a load-balancing strategy of data domain decomposition in parallel programming libraries for raster-based geocomputation based on the concept of spatial computational domain, which characterizes the distribution of computational intensity based on geocomputation characteristics. By implementing the proposed strategy with the message passing interface (MPI), a set of parallel raster-based geocomputation operators across different parallel computing platforms (known as PaRGO V2) was upgraded to improve load-balancing parallelization. The proposed strategy was evaluated by parallelizing two typical geocomputation algorithms (i.e. inverse distance weight interpolation and fuzzy c-means clustering) using PaRGO V2 with uneven distributed computational intensity. The results show that the proposed strategy with PaRGO V2, compared with the previously adopted data domain decomposition strategy, yielded significant improvements to the load balance (i.e. better parallel performance).
登录
查看更多内容
DOI：
10.1016/0167-739x(94)00063-k
发表时间：
1995-03
期刊：
Future Gener. Comput. Syst.
影响因子：
--
作者：
R. Bruce;S. Chapple;N. B. MacDonald;A. Trew;Shari Trewin
通讯作者：
R. Bruce;S. Chapple;N. B. MacDonald;A. Trew;Shari Trewin
DOI：
10.1109/geoinformatics.2012.6270298
发表时间：
2012-06
期刊：
2012 20th International Conference on Geoinformatics
影响因子：
--
作者：
Zhibing Jin;Yingxia Pu;Jie-chen Wang;Jingsong Ma;Gang Chen
通讯作者：
Zhibing Jin;Yingxia Pu;Jie-chen Wang;Jingsong Ma;Gang Chen
DOI：
10.1007/978-3-642-29737-3_31
发表时间：
2011-08
期刊：
--
影响因子：
--
作者：
Richard Membarth;Anton Lokhmotov;J. Teich
通讯作者：
Richard Membarth;Anton Lokhmotov;J. Teich
DOI：
10.1080/13658816.2013.771740
发表时间：
2013-11
影响因子：
5.7
作者：
Eric Shook;Shaowen Wang;Wenwu Tang
通讯作者：
Eric Shook;Shaowen Wang;Wenwu Tang
利用高性能计算应对空间大数据土地利用/土地覆盖变化分析的挑战
影响因子：
3.4
作者：
亢晓琛;刘纪平;董春;徐胜华
通讯作者：
徐胜华