Impact of data layouts on the efficiency of GPU-accelerated IDW interpolation

Mei, Gang; Tian, Hong

doi:10.1186/s40064-016-1731-6

Cited by 37 publications

(17 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(2) We employ the local set of data points to compute the prediction value of the interpolated point using different interpolation methods. Mei & Tian (2016) evaluated the impact of different data layouts on the computational efficiency of the GPU-accelerated IDW interpolation algorithm. They implemented three IDW versions of GPU implementations, based upon five data layouts, including the Structure of Arrays (SoA), the Array of Structures (AoS), the Array of aligned Structures (AoaS), the Structure of Arrays of aligned Structures (SoAoS), and a hybrid layout, then they carried out several groups of experiments to evaluate the impact of different data layouts on the interpolation efficiency.…”

Section: Implementations Of the Spatial Interpolation Algorithmsmentioning

confidence: 99%

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

Yang

et al. 2020

PeerJ Computer Science

Self Cite

View full text Add to dashboard Cite

The building of large-scale Digital Elevation Models (DEMs) using various interpolation algorithms is one of the key issues in geographic information science. Different choices of interpolation algorithms may trigger significant differences in interpolation accuracy and computational efficiency, and a proper interpolation algorithm needs to be carefully used based on the specific characteristics of the scene of interpolation. In this paper, we comparatively investigate the performance of parallel Radial Basis Function (RBF)-based, Moving Least Square (MLS)-based, and Shepard’s interpolation algorithms for building DEMs by evaluating the influence of terrain type, raw data density, and distribution patterns on the interpolation accuracy and computational efficiency. The drawn conclusions may help select a suitable interpolation algorithm in a specific scene to build large-scale DEMs.

show abstract

Section: Implementations Of the Spatial Interpolation Algorithmsmentioning

confidence: 99%

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

Yang

et al. 2020

PeerJ Computer Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…Typically, there are two major choices of the data layout: the array of structures (AoS) and the structure of arrays (SoA) [ 47 ]; another type of data layout, array of aligned structures (AoaS) [ 34 ], can be very easily generated by adding the forced alignment based on the layout AoS. In fact, the data layout AoaS can be considered as an improved variant of the layout AoS; see these three layouts, i.e.…”

Section: Graphics Processing Unit-accelerated Adaptive Inverse Distanmentioning

confidence: 99%

“…And quite recently, Mei [ 33 ] developed two GPU implementations of the IDW interpolation algorithm, the tiled version and the CUDA Dynamic Parallelism (CDP) version, by taking advantage of shared memory and CUDA Dynamic Parallelism, and found that the tiled version has speed-ups of 120 and 670 over the CPU version when the power parameter p was set to 2 and 3.0, respectively, but the CDP version is 4.8–6.0 times slower than the naive GPU version. In addition, Mei & Tian [ 34 ] compared and analysed the impact of data layouts on the efficiency of GPU-accelerated IDW implementations.…”

Section: Introductionmentioning

confidence: 99%

Accelerating adaptive inverse distance weighting interpolation algorithm on a graphics processing unit

Mei¹,

Xu²,

Xu³

2017

R. Soc. open sci.

Self Cite

View full text Add to dashboard Cite

This paper focuses on designing and implementing parallel adaptive inverse distance weighting (AIDW) interpolation algorithms by using the graphics processing unit (GPU). The AIDW is an improved version of the standard IDW, which can adaptively determine the power parameter according to the data points’ spatial distribution pattern and achieve more accurate predictions than those predicted by IDW. In this paper, we first present two versions of the GPU-accelerated AIDW, i.e. the naive version without profiting from the shared memory and the tiled version taking advantage of the shared memory. We also implement the naive version and the tiled version using two data layouts, structure of arrays and array of aligned structures, on both single and double precision. We then evaluate the performance of parallel AIDW by comparing it with its corresponding serial algorithm on three different machines equipped with the GPUs GT730M, M5000 and K40c. The experimental results indicate that: (i) there is no significant difference in the computational efficiency when different data layouts are employed; (ii) the tiled version is always slightly faster than the naive version; and (iii) on single precision the achieved speed-up can be up to 763 (on the GPU M5000), while on double precision the obtained highest speed-up is 197 (on the GPU K40c). To benefit the community, all source code and testing data related to the presented parallel AIDW algorithm are publicly available.

show abstract

“…The generality of the collision detection methods can be applied to numerous fields as it is a commonly encountered computational problem. Furthermore our strategies for the GPU have been applied in areas outside DEM [16,17].…”

Section: Impactmentioning

confidence: 99%

Blaze-DEMGPU: Modular high performance DEM framework for the GPU architecture

2016

View full text Add to dashboard Cite

Blaze-DEMGPU is a modular GPU based discrete element method (DEM) framework that supports polyhedral shaped particles. The high level performance is attributed to the light weight and Single Instruction Multiple Data (SIMD) that the GPU architecture offers. Blaze-DEMGPU offers suitable algorithms to conduct DEM simulations on the GPU and these algorithms can be extended and modified. Since a large number of scientific simulations are particle based, many of the algorithms and strategies for GPU implementation present in Blaze-DEMGPU can be applied to other fields. Blaze-DEMGPU will make it easier for new researchers to use high performance GPU computing as well as stimulate wider GPU research efforts by the DEM community.

show abstract

Impact of data layouts on the efficiency of GPU-accelerated IDW interpolation

Cited by 37 publications

References 18 publications

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

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

Accelerating adaptive inverse distance weighting interpolation algorithm on a graphics processing unit

Blaze-DEMGPU: Modular high performance DEM framework for the GPU architecture

Contact Info

Product

Resources

About