https://orcid.org/0000-0002-1182-0958
![Sample our Computer Science journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5928/TOC-Subject-Sample-2021-Computer-Science.jpg"})
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).
Acknowledgements
The authors sincerely thank the three anonymous reviewers for their constructive comments. The authors also thank Editor Dr. May Yuan for her kindness and patience during processing our submission. This work was supported by the Chinese Academy of Sciences under Grant XDA23100503, and the National Natural Science Foundation of China under Grant 41871362.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Data availability statement
PaRGO V2, which implemented the proposed strategy, is open source and available through GitHub (https://github.com/lreis2415/PaRGO). The experimental data together with PaRGO V2 are also available through a sharing link at https://doi.org/10.6084/m9.figshare.15345921.v1.
Additional information
Funding
Notes on contributors
Yu-Jing Wang
Yu-Jing Wang is a Ph.D. student at the State Key Laboratory of Resources & Environmental Information System, Institute of Geographical Sciences & Natural Resources Research, Chinese Academy of Science. His research interests include parallel geocomputation and watershed modeling and scenario analysis. In this study, he developed the program, conducted the evaluation experiments, and wrote the draft of this paper.
Bei-Bei Ai
Bei-Bei Ai received her Bachelor’s degree in geographic information systems from Wuhan University and Master’s degree in Geographic Information Science from the Institute of Geographical Sciences & Natural Resources Research, Chinese Academy of Science. Her research interests include parallel geocomputation. In this study, she designed the proposed strategy, developed the first version of the program, and conducted the primary experiment.
Cheng-Zhi Qin
Cheng-Zhi Qin is a Professor of GIScience in the State Key Laboratory of Resources & Environmental Information System, Institute of Geographical Sciences & Natural Resources Research, Chinese Academy of Science. His research interests include digital terrain analysis, watershed modeling and scenario analysis, intelligent geographical modeling, and parallel geocomputation. He conceived of the original idea of this research, acquired funding for this research, designed the experiments, and revised all versions of this paper.
A-Xing Zhu
A-Xing Zhu is a Professor of Geography in the Department of Geography, University of Wisconsin-Madison. His research focuses on geoinformatics, remote sensing, soil mapping, climatology, artificial intelligence, fuzzy logic, watershed modeling and scenario analysis, and intelligent geocomputation. In this study, he contributed to the experimental design and manuscript writing.