GPU-accelerated simulation of massive spatial data based on the modified planar rotator model

DOI10.1007/S11004-019-09835-3MaRDI QIDQ2284097zbMATH OpenOpenAlexWikidataFDO

Authors M. Žukovič, Michal Borovský, Matúš Lach, Dionissios T. Hristopulos

Publication date 14 January 2020

Published in Mathematical Geosciences (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1811.01604

zbMATH Keywords

conditional simulationCUDAGPU parallel computinghybrid Monte Carlonon-Gaussian modelspatial interpolation

Mathematics Subject Classification ID

Inference from spatial processes (62M30) Applications of statistics to physics (62P35) Parallel numerical computation (65Y05)

Abstract: A novel Gibbs Markov random field for spatial data on Cartesian grids based on the modified planar rotator (MPR) model of statistical physics has been recently introduced for efficient and automatic interpolation of big data sets, such as satellite and radar images. The MPR model does not rely on Gaussian assumptions. Spatial correlations are captured via nearest-neighbor interactions between transformed variables. This allows vectorization of the model which, along with an efficient hybrid Monte Carlo algorithm, leads to fast execution times that scale approximately linearly with system size. The present study takes advantage of the short-range nature of the interactions between the MPR variables to parallelize the algorithm on graphics processing units (GPU) in the Compute Unified Device Architecture (CUDA) programming environment. It is shown that, for the processors employed, the GPU implementation can lead to impressive computational speedups, up to almost 500 times on large grids, compared to the single-processor calculations. Consequently, massive data sets comprising millions of data points can be automatically processed in less than one second on an ordinary GPU.

Recommendations

• Simulating spatially explicit problems on high performance architectures
• Computing framework of large-scale dilated polyhedral DEM particles simulation based on GPU
• A CUDA-based implementation of an improved SPH method on GPU
• GPU-based parallel reservoir simulators
• Data structures and transformations for physically based simulation on a GPU

Cites work

• Analytic Properties and Covariance Functions for a New Class of Generalized Gibbs Random Fields
• Covariance tapering for likelihood-based estimation in large spatial data sets
• Equation of state calculations by fast computing machines
• Fast kriging of large data sets with Gaussian Markov random fields
• Fixed Rank Kriging for Very Large Spatial Data Sets
• Gaussian Markov Random Fields
• scientific article; zbMATH DE number 1934836 (Why is no real title available?)
• Multivariate spatial modeling for geostatistical data using convolved covariance functions
• On modelling discrete geological structures as Markov random fields
• Performance potential for simulating spin models on GPU
• Practical calculation of non-Gaussian multivariate moments in spatiotemporal Bayesian maximum entropy analysis
• Spartan Gibbs Random Field Models for Geostatistical Applications
• The probability perturbation method: a new look at Bayesian inverse modeling

Cited in

(2)

• Spatial data modeling by means of Gibbs-Markov random fields based on a generalized planar rotator model
• Development of a GPU-based high-performance radiative transfer model for the infrared atmospheric sounding interferometer (IASI)

Describes a project that uses

Uses Software

• CUDA
• GMRFLib

This page was built for publication: GPU-accelerated simulation of massive spatial data based on the modified planar rotator model

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2284097)