OpenCL Implementation of a Parallel Universal Kriging Algorithm for Massive Spatial Data Interpolation on Heterogeneous Systems

Fang Huang,Jian Tao,Xicheng Tan,Shuanshuan Bu

doi:10.3390/ijgi5060096

Abstract

In some digital Earth engineering applications, spatial interpolation algorithms are required to process and analyze large amounts of data. Due to its powerful computing capacity, heterogeneous computing has been used in many applications for data processing in various fields. In this study, we explore the design and implementation of a parallel universal kriging spatial interpolation algorithm using the OpenCL programming model on heterogeneous computing platforms for massive Geo-spatial data processing. This study focuses primarily on transforming the hotspots in serial algorithms, i.e., the universal kriging interpolation function, into the corresponding kernel function in OpenCL. We also employ parallelization and optimization techniques in our implementation to improve the code performance. Finally, based on the results of experiments performed on two different high performance heterogeneous platforms, i.e., an NVIDIA graphics processing unit system and an Intel Xeon Phi system (MIC), we show that the parallel universal kriging algorithm can achieve the highest speedup of up to 40× with a single computing device and the highest speedup of up to 80× with multiple devices.

Highlights

Spatial Interpolation (SI) is a process employed to estimate the values of properties at unknown points within an area covered by existing observed points [1]
SI is important for prediction and representation in many fields, including geographical information systems and remote sensing [4,5,6], geology [7], mining [8], hydrogeology [9], soil research [10], geophysics [11], oceanography [12], meteorology [13], ecology and environmental studies [14,15]
In contrast to other commonly-used SI algorithms, such as Voronoi and the inverse distance weighting method [36], it considers the spatial correlation between the points that need to be interpolated and their neighboring points, as well as giving the estimation error

Summary

Introduction

Spatial Interpolation (SI) is a process employed to estimate the values of properties at unknown points within an area covered by existing observed points [1]. Several different types of classification methods are used by SI procedures, e.g., point-area, global-local and exact-approximate interpolation [16]. The universal kriging interpolation algorithm is a type of linear and unbiased optimal kriging SI algorithm, which is used widely in many scientific and engineering applications. The universal kriging algorithm is a type of linear unbiased optimal SI algorithm. In contrast to other commonly-used SI algorithms, such as Voronoi and the inverse distance weighting method [36], it considers the spatial correlation between the points that need to be interpolated and their neighboring points, as well as giving the estimation error. The universal kriging algorithm provides more accurate interpolation results, and it is applied widely in the geological interpolation area. When the expectation of random variable Z pxq is a variable in the area of interest, we have,

Objectives

Methods

Results