Speeding up the high-accuracy surface modelling method with GPU

Abstract

In order to find a solution for accurate, topographic data-demanding applications, such as catchment hydrologic modeling and assessments of anthropic activities impact on environmental systems, high-accuracy surface modeling (HASM) method is developed. Although it can produce a digital elevation model (DEM) surface of higher accuracy than classical methods, e.g. inverse distance weighted, spline and kriging, HASM requires numerous iterations to solve large linear systems, which impede its applications in high-resolution and large-scale surface interpolation. This paper aims to demonstrate the utilization of graphics' processing units (GPUs) device to accelerate HASM in constructing large-scale and highresolution DEM surfaces. We parallelized the linear system algorithm for solving HASM with Compute Unified Device Architecture, a parallel programming model developed by NVIDIA. We designed a memory-saving strategy to enable the HASM algorithm to run on GPUs. The speedup ratio of GPU-based algorithm was tested and compared with CPUbased algorithm through simulations of both ideal Gaussian synthetic surface and real topographic surface in the loess plateau of Gansu province. The GPU-parallelized algorithm can attain an over 109 speedup ratio with the CPUbased algorithm as a reference. The speedup ratio increases with the scale and resolution of the dataset. The memory management strategy efficiently reduces the memory usage by more than eight times the grid cell number. Implementing HASM in the GPUs device enables modeling large-scale and high-resolution surfaces in a reasonable time period and implies the potential benefits from the use of GPUs as massive, parallel co-processors for arithmeticintensive data-processing applications.

A review of recent developments in HASM

“…GPU-based parallel algorithm of PCG can considerably improve the efficiency and robustness of HASM (Yan and Yue 2012a, b). GPU implementation of PCG algorithm for HASM is up to 12 times faster comparing with HASM-PCG (Yan et al 2015). However, it is not faster enough for many applications, especially at high temporal and spatial resolutions on the global level.…”

SURAA: A Novel Method and Tool for Loadbalanced and Coalesced SpMV Computations on GPUs

“…As noted earlier, WPK1 and WPK2 were unable to execute some of the matrices in the benchmark dataset and their results have not been included in computing mean and other qualities in the table. In order to calculate the aggregate performance of SURAA, we calculate aggregate speedup as in Equation (5). Figure 8 indicates the aggregate GFLOP/s achieved for SURAA and other tools.…”

“…Sparse Linear algebra is vital to scientific computations and various fields of engineering and thus has been included among the seven dwarfs [1] by the Berkeley researchers. Among the sparse numerical techniques, iterative solutions of sparse linear equation systems can be considered as of prime importance due to its application in various important areas such as solving finite differences of partial differential equations (PDEs) [2][3][4], high accuracy surface modelling [5], finding steady-state and transient solutions of Markov chains [6][7][8], probabilistic model checking [9][10][11], solving the time-fractional Schrödinger equation [12], web ranking [13][14][15], inventory control and manufacturing systems [16], queuing systems [17][18][19][20][21][22][23], fault modelling, weather forecasting, stochastic automata networks [24,25], communication systems and networks [26][27][28][29][30][31], reliability analysis [32], wireless and sensor networks [33][34][35][36][37], computational biology [38], healthcare [27,39,40], transportation [41,…”

Comparative investigation of parallel spatial interpolation algorithms for building large-scale digital elevation models

“…For example, Zhou et al (2017) proposed a parallel Open Multi-Processing (OpenMP)-and Message Passing Interface (MPI)-based implementation of the Priority-Flood algorithm that identifies and fills depressions in raster DEMs. Yan et al (2015) accelerated high-accuracy surface modeling (HASM) in constructing large-scale and fine resolution DEM surfaces by the use of GPUs and applied this acceleration algorithm to simulations of both ideal Gaussian synthetic surfaces and real topographic surfaces in the loess plateau of Gansu province. Tan et al (2017) presented a novel method to generate contour lines from grid DEM data, based on the programmable GPU pipeline, that can be easily integrated into a 3D GIS system.…”