Computational Technology Resources - CCP

Keywords: computational fluid dynamics, high-performance computing, high-order spectral method, domain decomposition, multilevel parallelism, arterial flow.

Summary

and methodologies in designing effective scalable solvers. We anticipate that in addition to hierarchical or multi-level domain decompositions, the major effort will be invested into the design of task-parallel codes and robust interface conditions.

In this paper we present one approach in building a scalable solver NekTarG for solution of multi-scale and large size problems [1]. NekTarG has been designed for multi-scale blood modeling. The macro-vascular scales describing the flow dynamics in large vessels are coupled to the mesovascular scales unfolding dynamics of individual blood-cells. The meso-vascular events are coupled to the microvascular ones accounting for blood perfusion, clot formation, adhesion of the blood cells to the arterial wall, etc.

The scalable-solver NekTarG integrates already existing, well tested and optimized codes (modules). Each module is targeting a single scale flow problem, i.e. at the macro-vascular scale, meso- or even micro-vascular scales. The modules are coupled using the multi-level communicating interface (MCI). The MCI has been designed for data-parallel and task-parallel decompositions, it minimizes the inter-module communication, and as such is also appropriate for grid computing.

To verify the feasibility of our approach and test the performance of NekTarG we solved a 8.2B degree of freedom benchmark flow problem on upto 122,800 cores. NekTarG has been also employed in simulations of blood flow in very large network of the brain arteries with an aneurysm. The good weak and strong scaling of the code obtained in benchmark tests as well as in arterial flow simulations in the brain are presented. We also describe computational methods of coupling a continuum high-order spectral/hp element deterministic solver to a stochastic solver DPD-LAMMPS, which is based on the dissipative particle dynamics method and present our first results.

References

1: L. Grinberg, G.E. Karniadakis, "A new Domain Decomposition Method With Overlapping Patches for Ultrascale Simulations: Application to Biological Flows", Journal of Computational Physics, 229(15), 5541-5563, 2010. doi:10.1016/j.jcp.2010.04.014

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 £85 +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: 95 PROCEEDINGS OF THE SECOND INTERNATIONAL CONFERENCE ON PARALLEL, DISTRIBUTED, GRID AND CLOUD COMPUTING FOR ENGINEERING Edited by: Paper 91 Parallel Paradigm for Ultraparallel Multi-Scale Brain Blood Flow Simulations L. Grinberg and G.E. Karniadakis Division of Applied Mathematics, Brown University, Providence RI, United States of America doi:10.4203/ccp.95.91 purchase the full-text of this paper Full Bibliographic Reference for this paper L. Grinberg, G.E. Karniadakis, "Parallel Paradigm for Ultraparallel Multi-Scale Brain Blood Flow Simulations", in , (Editors), "Proceedings of the Second International Conference on Parallel, Distributed, Grid and Cloud Computing for Engineering", Civil-Comp Press, Stirlingshire, UK, Paper 91, 2011. doi:10.4203/ccp.95.91 Keywords: computational fluid dynamics, high-performance computing, high-order spectral method, domain decomposition, multilevel parallelism, arterial flow. Summary and methodologies in designing effective scalable solvers. We anticipate that in addition to hierarchical or multi-level domain decompositions, the major effort will be invested into the design of task-parallel codes and robust interface conditions. In this paper we present one approach in building a scalable solver NekTarG for solution of multi-scale and large size problems [1]. NekTarG has been designed for multi-scale blood modeling. The macro-vascular scales describing the flow dynamics in large vessels are coupled to the mesovascular scales unfolding dynamics of individual blood-cells. The meso-vascular events are coupled to the microvascular ones accounting for blood perfusion, clot formation, adhesion of the blood cells to the arterial wall, etc. The scalable-solver NekTarG integrates already existing, well tested and optimized codes (modules). Each module is targeting a single scale flow problem, i.e. at the macro-vascular scale, meso- or even micro-vascular scales. The modules are coupled using the multi-level communicating interface (MCI). The MCI has been designed for data-parallel and task-parallel decompositions, it minimizes the inter-module communication, and as such is also appropriate for grid computing. To verify the feasibility of our approach and test the performance of NekTarG we solved a 8.2B degree of freedom benchmark flow problem on upto 122,800 cores. NekTarG has been also employed in simulations of blood flow in very large network of the brain arteries with an aneurysm. The good weak and strong scaling of the code obtained in benchmark tests as well as in arterial flow simulations in the brain are presented. We also describe computational methods of coupling a continuum high-order spectral/hp element deterministic solver to a stochastic solver DPD-LAMMPS, which is based on the dissipative particle dynamics method and present our first results. References 1 L. Grinberg, G.E. Karniadakis, "A new Domain Decomposition Method With Overlapping Patches for Ultrascale Simulations: Application to Biological Flows", Journal of Computational Physics, 229(15), 5541-5563, 2010. doi:10.1016/j.jcp.2010.04.014 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 £85 +P&P)
Back to top	©Civil-Comp Limited 2023 - terms & conditions