面向大规模空间Agent建模的分布式地理模拟框架

doi:10.12082/dqxxkx.2022.220001

摘要/Abstract

摘要：

基于Agent建模的地理模拟是认识和理解动态地理现象的有效方法,但随着地理模拟的规模和复杂性不断增加,模型的计算问题开始凸显。分布式并行仿真是解决大规模Agent复杂模拟计算的途径,然而已有研究基于Agent建模/仿真软件构建并行仿真系统的方式并不适用于具有高移动与行为交互的空间Agent建模及其模拟过程的实时可视化。为解决这个问题, 本文提出了一个分布式地理模拟框架DGSimF,用于大规模动态空间Agent模拟,支持模拟过程的实时表示与分析。设计了一个简单但高效的时空数据模型建模空间Agent,支持直接基于Agent行为建模集成地学模型,采用了时间微分方法协同各计算节点行为的执行,实现以“任务并行”的方式进行分布式计算以提高仿真性能,构建了基于三维地球渲染引擎的虚拟地理环境,提供模拟过程的实时可视化。最后,以“红蓝对抗”案例进行了实验验证,对不同模拟计算量和不同客户端数量下的仿真性能进行了分析,结果表明DGSimF可以为具有时空特征变化与行为交互的大规模空间Agent模拟提供一个有效的平台。通过扩展计算节点,DGSimF可以有效地缓解复杂模拟计算的压力问题,并且仿真性能较高,在实验中并行效率保持在0.7以上。

关键词: 地理模拟, 空间Agent, 分布式计算, 并行仿真, 基于Agent建模, 行为建模, 虚拟地理环境, 实时可视化

Abstract:

Geospatial simulation based on agent-based modeling is an effective method to recognize and understand dynamic geographic phenomena. As the scale and complexity of geospatial simulation continues to increase, the challenges in model computation increase. Distributed parallel simulation could be used to solve the complex simulation issue of large-scale agents. However, the existing research on building parallel simulation system based on agent modeling/simulation software is not suitable for modeling of spatial agent with high-mobility and behavioral interaction with others, and real-time visualization of simulation process. To solve this problem, this paper proposes a distributed geospatial simulation framework, namely DGSimF, for massive dynamic spatial agent modeling, which supports real-time representation and analysis of the simulation process. A simple but efficient spatial modeling agent for spatial-temporal data is designed, which supports the modeling of integrated geoscience models directly based on agent behavior, adopts the time differentiation method to coordinate the execution of the behavior of each computing node, supports distributed computation in the way of "task parallel" to improve the simulation performance, and constructs a Virtual Geographic Environment (VGE) based on three-dimensional earth rendering engine to support real-time visualization of intermediate simulation results. Finally, the experiments based on the "Red vs. Blue" case are carried out, and the simulation performance with different computation cost and different number of clients is analyzed. The results show that DGSimF can provide an effective platform for massive spatial agent simulation of spatio-temporal feature change and behavior interaction. By expanding the computing nodes, the pressure of complex simulation calculation can be effectively alleviated. Meanwhile, the simulation performance of the proposed framework is high, and the parallel efficiency remains above 0.7 in these experiments.

Key words: geospatial simulation, spatial agent, distributed computing, parallel simulation, Agent-based modeling, behavior modeling, virtual geographic environment, real-time visualization

曾梦熊, 华一新, 张政, 张江水, 杨振凯, 韦原原. 面向大规模空间Agent建模的分布式地理模拟框架[J]. 地球信息科学学报, 2022, 24(5): 815-826.DOI:10.12082/dqxxkx.2022.220001

ZENG Mengxiong, HUA Yixin, ZHANG Zheng, ZHANG Jiangshui, YANG Zhenkai, WEI Yuanyuan. A Distributed Geospatial Simulation Framework for Massive Spatial Agent-Based Modeling[J]. Journal of Geo-information Science, 2022, 24(5): 815-826.DOI:10.12082/dqxxkx.2022.220001

图/表 12

图1

图2

图3

图4

图5

图6

图7

表1

表2

表3

图8

表4

参考文献 28

[1]	Wallentin G. Spatial simulation: A spatial perspective on individual-based ecology—a review[J]. Ecological Modelling, 2017, 350(4):30-41. DOI: 10.1016/j.ecolmodel.2017.01.017 doi: 10.1016/j.ecolmodel.2017.01.017
[2]	余强毅, 吴文斌, 唐华俊, 等. 复杂系统理论与Agent模型在土地变化科学中的研究进展[J]. 地理学报, 2011, 66(11):1518-1530.
	[ Yu Q Y, Wu W B, Tang H J, et al. Complex system theory and Agent-based modeling: progresses in land change science[J]. Acta Geographica Sinica, 2011, 66(11):1518-1530. ] DOI: 10.11821/xb201111008 doi: 10.11821/xb201111008
[3]	翟瑞雪, 戴尔阜. 基于主体模型的人地系统复杂性研究[J]. 地理研究, 2017, 36(10):1925-1935. doi: 10.11821/dlyj201710009
	[ Zhai R X, Dai E F. Research on the complexity of man-land system based on agent-based models[J]. Geographical Research, 2017, 36(10):1925-1935. ] DOI: 10.11821/dlyj201710009 doi: 10.11821/dlyj201710009
[4]	Ding Y L, Zhu Q, Lin H. An integrated virtual geographic environmental simulation framework: A case study of flood disaster simulation[J]. Geo-spatial Information Science, 2014, 17(4):190-200. DOI: 10.1080/10095020.2014.988199 doi: 10.1080/10095020.2014.988199
[5]	Arifin S M N, Madey G R, Collins F H. Spatial agent-based simulation modeling in public health: design, implementation, and applications for malaria epidemiology[M]. New Jersey: John Wiley & Sons Inc, 2016. DOI: 10.1002/9781118964385 doi: 10.1002/9781118964385
[6]	潘茂林, 陶海燕. JAVA-SWARM与GIS集成研究——以城市地理模拟为例[J]. 系统仿真学报, 2009, 21(18):5704-5708.
	[ Pan M L, Tao H Y. Integration technique between java-swarm and GIS on urban geo-simulation[J]. Journal of System Simulation, 2009, 21(18):5704-5708. ] DOI: CNKI:SUN:XTFZ.0.2009-18-025 doi: CNKI:SUN:XTFZ.0.2009-18-025
[7]	黎夏, 李丹, 刘小平. 地理模拟优化系统(GeoSOS)及其在地理国情分析中的应用[J]. 测绘学报, 2017, 46(10):1598-1608.
	[ Li X, Li D, Liu X P. Geographical simulation and optimization system(GeoSOS) and its appl ication in the analysis of geographic national conditions[J]. Acta Geodaetica et Cartographica Sinica, 2017,46(10):1598-1608. ]DOI: CNKI:SUN:DXJZ.0.2009-08-009 doi: CNKI:SUN:DXJZ.0.2009-08-009
[8]	Vahidnia M H, Alesheikh A A, Alavipanah S K. A multi-agent architecture for geosimulation of moving agents[J]. Journal of Geographical Systems, 2015, 17(4):353-390.DOI: 10.1007/s10109-015-0218-2 doi: 10.1007/s10109-015-0218-2
[9]	Sauvage S, Filatova T, Horstman E, et al. Modelling adaptive behaviour in spatial agent-based models: coastal cities and climate change [C]// International Congress on Environmental Modelling and Software, Toulouse, 2016: 18.
[10]	Li Z Q, Guan X F, Li R, et al. 4D-SAS: A distributed dynamic-data driven simulation and analysis system for massive spatial Agent-based modeling[J]. ISPRS International Journal of Geo-Information, 2016, 42(5):1-21. DOI: 10.3390/ijgi5040042 doi: 10.3390/ijgi5040042
[11]	Züfle A, Kavak H, Kim J S, et al. Towards large-scale Agent-based geospatial simulation [C]// Social Computing, Behavioral-Cultural Modeling and Prediction and Behavior Representation in Modeling and Simulation, Washington DC, 2021:1-10.
[12]	Lettieri N, Spagnuolo C, Vicidomini L. Distributed agent-based simulation and GIS: An experiment with the dynamics of social norms [C]// European Conference on Parallel Processing, Vienna, 2015:379-391. DOI: 10.1007/978-3-319-27308-231 doi: 10.1007/978-3-319-27308-231
[13]	Francisco B, Albert G-M, Emilio L, et al. Care HPS: a high performance simulation tool for parallel and distributed agent-based modeling[J]. Future Generation Computer Systems, 2017, 68(3):59-73. DOI: 10.1016/j.future.2016.08.015 doi: 10.1016/j.future.2016.08.015
[14]	Mastio M, Zargayouna M, Scémama G, et al. Distributed agent-based traffic simulations[J]. IEEE Intelligent Transportation Systems Magazine, 2018, 10(1):145-156. DOI: 10.1109/MITS.2017.2776162 doi: 10.1109/MITS.2017.2776162
[15]	Simon C, Marian G, Mike H, et al. Exploitation of high performance computing in the FLAME Agent-based simulation framework [C]// IEEE International Conference on High Performance Computing & Communication, Liverpool, 2012:538-545. DOI: 10.1109/HPCC.2012.79 doi: 10.1109/HPCC.2012.79
[16]	Gennaro C, Carmine S, Vittorio S. Toward the new version of D-MASON: efficiency, effectiveness and correctness in parallel and distributed agent-based simulations [C]// IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW), Chicago, 2016:1803-1812. DOI: 10.1109/IPDPSW.2016.52 doi: 10.1109/IPDPSW.2016.52
[17]	Marc J-R, Jan K, Martin B. A coupled simulation architecture for agent-based/geohydrological modelling with NetLogo and MODFLOW[J]. Environmental Modelling and Software, 2019, 115(3):19-37. DOI: 10.1016/j.envsoft.2019.01.020 doi: 10.1016/j.envsoft.2019.01.020
[18]	赵耀龙, 张珂, 彭永俊, 等. 基于地理模拟方法的昆明市空间拓展情景分析[J]. 地理研究, 2014, 33(1):119-131. doi: 10.11821/dlyj201401011
	[ Zhao Y L, Zhang K, Peng Y J, et al. Scenario analysis of urban growth in kunming based on geosimulation system[J]. Geographical Research, 2014, 33(1):119-131. ] DOI: 10.11821/dlyj201401011 doi: 10.11821/dlyj201401011
[19]	Lyuba M, Carole A A, Julie D. Multi-agent geospatial simulation of human interactions and behaviour in bushfires [C]// 16th International Conference on Information Systems for Crisis Response and Management, València, 2019:1-15.
[20]	Hua T, Jiang T. Agent-based modeling and simulation of complex distributed systems [C]// World Congress on Intelligent Control & Automation, Chongqing, 2017:416-421. DOI: 10.1109/WCICA.2008.4592960 doi: 10.1109/WCICA.2008.4592960
[21]	Andreas K. Geosimulation: modeling spatial processes[M]. Berlin: Springer, 2015. DOI: 10.1007/978-3-319-00008-4_5 doi: 10.1007/978-3-319-00008-4_5
[22]	林珲, 胡明远, 陈旻, 等. 从地理信息系统到虚拟地理环境的认知转变[J]. 地球信息科学学报, 2020, 22(4):662-672. doi: 10.12082/dqxxkx.2020.200048
	[ Lin H, Hu M Y, Chen M, et al. Cognitive transformation from geographic information system to virtual geographic environments[J]. Journal of Geo-information Science, 2020, 22(4):662-672. ] DOI: 10.12082/dqxxkx.2020.200048 doi: 10.12082/dqxxkx.2020.200048
[23]	林珲, 黄凤茹, 闾国年. 虚拟地理环境研究的兴起与实验地理学新方向[J]地理学报, 2009, 64(1):7-20.
	[ Lin H, Huang F R, Lv G N. Development of virtual geographic environments and the new initiative in experimental geography[J]. Acta Geographica Sinica, 2009, 64(1):7-20. ] DOI: 10.3321/j.issn:0375-5444.2009.01.002 doi: 10.3321/j.issn:0375-5444.2009.01.002
[24]	Mekni M. Integration of GIS data for visualization of virtual geospatial environments [C]// Second International Conference on Mathematics and Computers in Sciences and in Industry (MCSI), Sliema, 2015:273-282. DOI: 10.1109/MCSI.2015.39 doi: 10.1109/MCSI.2015.39
[25]	Mark D H, Michael R M. Amdahl's Law in the multicore era[J]. Computer, 2008, 41(7):33-38. DOI: 10.1109/HPCA.2008.4658638 doi: 10.1109/HPCA.2008.4658638
[26]	Dhananjai M R. Efficient parallel simulation of spatially-explicit agent-based epidemiological models[J]. Journal of Parallel and Distributed Computing, 2016, 93(C):102-119.DOI: 10.1016/j.jpdc.2016.04.004 doi: 10.1016/j.jpdc.2016.04.004
[27]	Omar R, Yann S, Philippe M. Load-balancing for large scale situated Agent-based simulations[J]. Procedia Computer Science, 2015, 51:90-99. DOI: 10.1016/j.procs.2015.05.204 doi: 10.1016/j.procs.2015.05.204
[28]	Petkova A, Hughes C, Deo N, et al. Accelerating the distributed simulations of agent-based models using community detection [C]// IEEE Rivf International Conference on Computing & Communication Technologies, Hanoi, 2016:25-30. DOI: 10.1109/RIVF.2016.7800264 doi: 10.1109/RIVF.2016.7800264

类型	STDB	DSIME	客户端#1	客户端#2	客户端#3	管理端
数量	1	3	1	1	1	1
CPU	Intel Xeon E5-2620(6×12 cores, 2.4GHz)	Intel Xeon E5-2620(6×12 cores,2.4 GHz)	Intel Core i7-6700(4×8 cores,3.4 GHz)	Intel Core i7-7700HQ (4×8 cores, 2.8 GHz)	Intel Core i7-6700(4×8 cores, 3.40 GHz)	Intel Core i7-10750 (6×12 cores,2.6 GHz)
内存	32 GB (2133 MHz)	32 GB (2133 MHz)	8 GB (2133 MHz)	16 GB (2400 MHz)	16 GB (2933 MHz)	16 GB (3200 MHz)
硬盘	12 TB (10krpm,SAS)	12 TB (10krpm,SAS)	1 TB (7.5krpm,SAS)	1 TB (7.5krpm,SAS)	2 TB (7.5krpm,SAS)	1 TB (7.5krpm,SAS)
内核	3.10.0-327.el7.x86_64	3.10.0-327.el7.x86_64
操作系统	CentOS 7	CentOS 7	Windows 10	Windows 10	Windows 10	Windows 10

计算量/s	T(1)/s	T(2) /s	S(2)	E(2)
0.2	7083	4218	1.6792	0.8396
1	13 164	7461	1.7644	0.8822
3	24 912	12 711	1.9599	0.9799
5	37 470	19 101	1.9617	0.9808
7	49 932	25 629	1.9483	0.9741
9	61 488	31 473	1.9537	0.9768

计算量/s	T(1) /s	T(3) /s	S(3)	E(3)
0.2	7083	3309	2.1405	0.7135
3	24 912	10215	2.4388	0.8129
7	37 470	17 352	2.8776	0.9592
9	61 488	21 702	2.8333	0.9444

计算量/s	26 个空间Agent		1026 个空间Agent
计算量/s	S(2)	E(2)	S(2)	E(2)
0.2	1.0858	0.5429	1.6792	0.8396
1	1.2926	0.6463	1.7644	0.8822
3	1.4210	0.7105	1.9599	0.9799
5	1.5623	0.7811	1.9617	0.9808