A CUDA-based parallel accelerating geographically weighted regression algorithm for big data

doi:10.13474/j.cnki.11-2246.2020.0379

Abstract

Abstract: In order to improve calculation speed and reduce memory consumption in traditional GWR, this paper proposes a fast parallel geographically weighted regression (FPGWR) algorithm, which bases on the NVIDIA computed unified device architecture (CUDA) to achieve parallel acceleration of GWR. The FPGWR algorithm finely decomposes the serial process into parallel independent computing modules and optimizes the memory usage model to increase the speed of GWR algorithm. This paper compares the calculation speed of FPGWR and traditional GWR towards simulated and real data. Results show that FPGWR can support a larger sample calculation and effectively increase the calculation speed. In addition, the larger the amount of data, the faster the algorithm presents.

Key words: GWR, CUDA, GPU, parallel acceleration, big data

CLC Number:

LIU Zhentao, YANG Yi, WANG Dongchao, XIE Xiaoyao. A CUDA-based parallel accelerating geographically weighted regression algorithm for big data[J]. Bulletin of Surveying and Mapping, 2020, 0(12): 1-5.

References

[1] BRUNSDON C, FOTHERINGHAM A S, CHARLTON M E. Geographically weighted regression:a method for exploring spatial nonstationarity[J]. Geographical Analysis, 1996, 28(4):281-298.
[2] HU X, WALLER L A, AL-HAMDAN M Z, et al. Estimating ground-level PM2.5 concentrations in the southeastern US using geographically weighted regression[J]. Environmental Research, 2013, 121:1-10.
[3] KUMAR S, LAL R, LIU D. A geographically weighted regression kriging approach for mapping soilorganic carbon stock[J]. Geoderma, 2012, 189-190:627-634.
[4] SU S, XIAO R, ZHANG Y. Multi-scale analysis of spatially varying relationships between agricultural landscape patterns and urbanization using geographically weighted regression[J].Applied Geography, 2012, 32(2):360-375.
[5] 刘彦文,刘成武,何宗宜,等.基于地理加权回归模型的武汉城市圈生态用地时空演变及影响因素[J].应用生态学报,2020,31(3):987-998.
[6] HU M, LI Z, WANG J, et al. Determinants of the incidence of hand, foot and mouth disease in China using geographically weighted regression models[J].Plos One, 2012, 7(6):e38978.
[7] LU B, CHARLTON M, HARRIS P, et al. Geographically weighted regression with a non-Euclidean distance metric:a case study using hedonic house price data[J].International Journal of Geographical Information Science, 2014, 28(4):660-681.
[8] TRAN H T, NGUYEN H T, TRAN V T. Large-scale geographically weighted regression on Spark[C]//Eighth International Conference on Knowledge and Systems Engineering (KSE). Hanoi:IEEE, 2016:127-132.
[9] ZHANG J. Towards personal high-performance geospatial computing (HPC-G):perspectives and a case study[C]//Proceedings of the ACM SIGSPATIAL international workshop on high performance and distributed geographic information systems.[S.l.]:ACM, 2010:3-10.
[10] ZHANG J, YOU S. CudaGIS:report on the design and realization of a massive data parallel GIS on GPUs[C]//Proceedings of the 3rd ACM SIGSPATIAL International Workshop on GeoStreaming.[S.l.]:ACM, 2012:101-108.
[11] ZHANG J, YOU S, GRUENWALD L. Large-scale spatial data processing on GPUs and GPU-accelerated clusters[J]. Sigspatial Special, 2015, 6(3):27-34.
[12] FOSTER S A, GORR W L. An adaptive filter for estimating spatially-varying parameters:Application to modeling police hours spent in response to calls for service[J]. Management Science, 1986, 32(7):878-889.
[13] FOTHERINGHAM A S, CHARLTON M, BRUNSDON C. Measuring spatial variations in relationships with geographically weighted regression[M]//Recent developments in spatial analysis. Berlin:Springer, 1997:60-82.
[14] BRUNSDON C, FOTHERINGHAM S, CHARLTON M. Geographically weighted regression[J]. Journal of the Royal Statistical Society:Series D (The Statistician), 1998, 47(3):431-443.
[15] AKAIKE H. A new look at the statistical model identification[J]. IEEE transactions on automatic control, 1974, 19(6):716-723.
[16] 杨毅.顾及时空非平稳性的地理加权回归方法研究[D]. 武汉:武汉大学,2016.