Computational Technology Resources - CCP

Keywords: poly-dispersed particle systems, discrete element method, grid computing, gLite middleware, grid visualization e-service, VTK.

Summary

Efficient numerical investigation of challenging scientific and industrial problems is closely related to development of advanced computer systems. Among various physical problems simulation of particulate solids, grains and powders is a recent target of many industries, while the discrete element method (DEM) [1] employing inter-particle contacts is the dominating technology applied for the above purposes. Three-dimensional simulation and visualization of poly-dispersed particle systems is performed by the developed software on the BalticGrid [2] build by using gLite middleware.

An efficient parallel algorithm for moving particles, that exchange processors, is incorporated in the domain decomposition framework. Unblocked message passing inter-processor communication (MPI) is implemented in the code DEMMAT_PAR. The deployment of DEMMAT_PAR in the BalticGrid-II relies on the modern MPI-START system.

Two types of problems, tri-axial compacting of granular material as well as granular flow in hopper, were solved on the BalticGrid-II testbed. The densification phenomenon during compacting is characterised by time variation of various microscopic and macroscopic variables such as kinetic energy, packing density, coordination number and wall pressures. The granular flow from hoppers and silos has been an important problem investigated by using the DEM for many years. It has a wide range of applications in industry retaining, however, relative simplicity. Increasing computational capacities yields better quality for characterisation of granular flow as illustrated in the paper.

Parallel performance of the developed software was also investigated by a series of benchmark tests with different numbers of heterogeneous particles. The measured parallel efficiency showed that the implemented inter-processor communication algorithms were well designed for the distributed memory PC clusters. However, heterogeneity of material significantly influenced parallel performance of the code.

Visualization of poly-dispersed particle systems lacking inherent connections was performed in a remote grid environment. Novel visualization e-service VizLitG was developed for convenient access and efficient visualization of results produced by scientific computations on the grid. VizLitG has a flexible environment and remote instrumentation provided by Java EE and GlassFish [3]. Employed gVid software provides for the remote user with full interactivity level. Performance tests of VizLitG were made on datasets of different sizes. A HP workstation equipped with better hardware ran visulization tests faster than the PC considered, but the difference obtained was not very significant.

References

1: P.A. Cundall, O.D.L. Strack, "A discrete numerical model for granular assemblies", Geotechnique, 29(1), 47-65, 1979.
2: BalticGrid: http://www.balticgrid.eu
3: GlassFish: https://glassfish.dev.java.net

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £72 +P&P)

	Computational & Technology Resources an online resource for computational, engineering & technology publications
	not logged in - login
Front Page Browse CCP CSETS CTR IJRT Other Authors Search Purchase Guide FAQ Contact us	Civil-Comp Proceedings ISSN 1759-3433 CCP: 90 PROCEEDINGS OF THE FIRST INTERNATIONAL CONFERENCE ON PARALLEL, DISTRIBUTED AND GRID COMPUTING FOR ENGINEERING Edited by: Paper 20 Computation and Visualization of Poly-Dispersed Particle Systems on gLite Grid A. Kaceniauskas¹, R. Kacianauskas², A. Maknickas¹ and D. Markauskas² ¹Laboratory of Parallel Computing, ²Laboratory of Numerical Modelling, Vilnius Gediminas Technical University, Lithuania doi:10.4203/ccp.90.20 purchase the full-text of this paper Full Bibliographic Reference for this paper A. Kaceniauskas, R. Kacianauskas, A. Maknickas, D. Markauskas, "Computation and Visualization of Poly-Dispersed Particle Systems on gLite Grid", in , (Editors), "Proceedings of the First International Conference on Parallel, Distributed and Grid Computing for Engineering", Civil-Comp Press, Stirlingshire, UK, Paper 20, 2009. doi:10.4203/ccp.90.20 Keywords: poly-dispersed particle systems, discrete element method, grid computing, gLite middleware, grid visualization e-service, VTK. Summary Efficient numerical investigation of challenging scientific and industrial problems is closely related to development of advanced computer systems. Among various physical problems simulation of particulate solids, grains and powders is a recent target of many industries, while the discrete element method (DEM) [1] employing inter-particle contacts is the dominating technology applied for the above purposes. Three-dimensional simulation and visualization of poly-dispersed particle systems is performed by the developed software on the BalticGrid [2] build by using gLite middleware. An efficient parallel algorithm for moving particles, that exchange processors, is incorporated in the domain decomposition framework. Unblocked message passing inter-processor communication (MPI) is implemented in the code DEMMAT_PAR. The deployment of DEMMAT_PAR in the BalticGrid-II relies on the modern MPI-START system. Two types of problems, tri-axial compacting of granular material as well as granular flow in hopper, were solved on the BalticGrid-II testbed. The densification phenomenon during compacting is characterised by time variation of various microscopic and macroscopic variables such as kinetic energy, packing density, coordination number and wall pressures. The granular flow from hoppers and silos has been an important problem investigated by using the DEM for many years. It has a wide range of applications in industry retaining, however, relative simplicity. Increasing computational capacities yields better quality for characterisation of granular flow as illustrated in the paper. Parallel performance of the developed software was also investigated by a series of benchmark tests with different numbers of heterogeneous particles. The measured parallel efficiency showed that the implemented inter-processor communication algorithms were well designed for the distributed memory PC clusters. However, heterogeneity of material significantly influenced parallel performance of the code. Visualization of poly-dispersed particle systems lacking inherent connections was performed in a remote grid environment. Novel visualization e-service VizLitG was developed for convenient access and efficient visualization of results produced by scientific computations on the grid. VizLitG has a flexible environment and remote instrumentation provided by Java EE and GlassFish [3]. Employed gVid software provides for the remote user with full interactivity level. Performance tests of VizLitG were made on datasets of different sizes. A HP workstation equipped with better hardware ran visulization tests faster than the PC considered, but the difference obtained was not very significant. References 1 P.A. Cundall, O.D.L. Strack, "A discrete numerical model for granular assemblies", Geotechnique, 29(1), 47-65, 1979. 2 BalticGrid: http://www.balticgrid.eu 3 GlassFish: https://glassfish.dev.java.net purchase the full-text of this paper (price £20) go to the previous paper go to the next paper return to the table of contents return to the book description purchase this book (price £72 +P&P)
Back to top	©Civil-Comp Limited 2023 - terms & conditions