A component-based architecture for parallel multi-physics PDE simulation |
| |
| Affiliation: | 1. Research Team OASIS, INRIA Sophia Antipolis - Mediterranean, 2004 Routes des Lucioles, 06902 Sophia Antipolis, France;2. Polytech Nice Sophia, University of Nice - Sophia Antipolis, 930 Route des Colles - BP 145, 06903 Sophia Antipolis Cedex, France;3. Center for Research in Distributed & Supercomputing, Faculty of Computing, Riphah International University, I-14, Islamabad, Pakistan;1. Institute for Applied Materials, Karlsruhe Institute of Technology (KIT), Germany;2. School of Civil Engineering, The University of Sydney, Sydney, NSW, 2006, Australia;1. College of Information Sciences and Technology, Donghua University, Shanghai 201620, China;2. MRC—University of Glasgow Centre for Virus Research, Glasgow G11 5JR, United Kingdom;3. Engineering Research Center of Digitized Textile & Fashion Technology, Ministry of Education, Shanghai 201620, China;1. Department of Imaging and Interventional Radiology, The Chinese University of Hong Kong, Prince of Wales Hospital, Hong Kong, China;2. State Key Laboratory in Oncology in South China, Sir Y.K. Pao Center for Cancer, Department of Clinical Oncology, Hong Kong Cancer Institute and Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong, China;3. Imaging Department, St Mary''s Hospital, Imperial College Healthcare NHS Trust, London, UK;1. Department of Imaging and Interventional Radiology, The Chinese University of Hong Kong, Prince of Wales Hospital, Hong Kong Special Administrative Region;2. State Key Laboratory in Oncology in South China, Sir Y.K. Pao Center for Cancer, Department of Clinical Oncology, Hong Kong Cancer Institute and Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong Special Administrative Region;3. Imaging, Imperial College Healthcare NHS Trust, London, United Kingdom |
| |
| Abstract: | We describe the Uintah Computational Framework (UCF), a set of software components and libraries that facilitate the simulation of partial differential equations on structured adaptive mesh refinement grids using hundreds to thousands of processors. The UCF uses a non-traditional approach to achieving parallelism, employing an abstract taskgraph representation to describe computation and communication. This representation has a number of advantages that affect the performance of the resulting simulation. We demonstrate performance of the system on a solid mechanics algorithm, two different computational fluid-dynamics (CFD) algorithms, as well as coupled CFD/mechanics algorithms. We show performance of the UCF using up to 2000 processors. |
| |
| Keywords: | |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
