A load balancing method in accelerating Kriging algorithm on CPU-GPU heterogeneous platforms

被引：0

作者：

Jiang, Chunlei ^{[1
,2
]}

Zhang, Shuqing ^{[1
]}

机构：

[1] Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences, Changchun

[2] University of Chinese Academy of Sciences, Beijing

来源：

Guofang Keji Daxue Xuebao/Journal of National University of Defense Technology | 2015年 / 37卷 / 05期

关键词：

General purpose graphics processor units; Kriging interpolation; Load balancing; Open computing language;

D O I：

10.11887/j.cn.201505006

中图分类号：

学科分类号：

摘要：

Kriging interpolation algorithm is of great practical significance and is widely applied to various fields of geoscience. However, Kriging interpolation would inevitably encounter the performance bottleneck when the output grid or input samples increase. Implemented with OpenCL and OpenMP, the ordinary Kriging interpolation was accelerated on heterogeneous platforms: GPU and CPU. By considering the performance difference of CPU and GPU on the densities of samples, a new load balancing method of LBCPDD (Load Balancing based on Computation Performance and Data Distribution) was proposed, in which not only hardware performance but also data distribution characteristics were taken into account. Experiment results show that LBCPDD method can effectively enhance the speed of ordinary Kriging, save memory space and improve the efficiency of memory access. ©, 2015, National University of Defense Technology. All right reserved.

引用

页码：35 / 39and148

共 15 条

[1]

Krige D.G., A statistical approach to some basic mine valuation problems on the witwatersrand, Journal of the Chemical Metallurgical & Mining Society of South Africa, 94, 3, pp. 95-111, (1951)

[2]

de Baar J.H.S., Dwight R.P., Bijl H., Speeding up Kriging through fast estimation of the hyperparameters in the frequency-domain, Computers & Geosciences, 54, pp. 99-106, (2013)

[3]

Wu L., Yang Y., Qin C., Et al., On basic geographic parallel algorithms of new generation GIS for new hardware architectures, Geography and Geo-Information Science, 29, 4, pp. 1-8, (2013)

[4]

Cheng G., Chen L., Wu Q., Et al., A task scheduling method for parallelization of complicated geospatial raster data processing algorithms, Journal of National University of Defense Technology, 34, 6, pp. 61-65, (2012)

[5]

Hu H.D., Shu H., An improved coarse-grained parallel algorithm for computational acceleration of ordinary Kriging interpolation, Computers & Geosciences, 78, pp. 44-52, (2015)

[6]

Cheng T.P., Li D.D., Wang Q., On parallelizing universal Kriging interpolation based on OpenMP, Proceedings of the 9th International Symposium on Distributed Computing and Applications to Business, Engineering and Science, pp. 36-39, (2010)

[7]

Cheng T.P., Accelerating universal Kriging interpolation algorithm using CUDA-enabled GPU, Computers & Geosciences, 54, pp. 178-183, (2013)

[8]

Shi X., Ye F., Kriging interpolation over heterogeneous computer architectures and systems, GIScience & Remote Sensing, 50, 2, pp. 196-211, (2013)

[9]

Lu F., Song J., Yin F., Et al., Survey of CPU/GPU synergetic parallel computing, Computer Science, 38, 3, pp. 5-9, (2011)

[10]

Fang L., Wang M., Li D., Et al., A workload-distribution based CPU/GPU MTF compensation approach for high resolution satellite images, Acta Geodaetica et Cartographica Sinica, 43, 6, pp. 598-606, (2014)

← 1 2 →