Filters








2,870 Hits in 4.4 sec

Massively parallel implementation of a 3D vortex-boundary element method

Adrin Gharakhani, Ahmed F. Ghoniem, Yves Gagnon, Georges-Henri Cottet, David G., Ahmed F., Eckart Meiburg
1996 ESAIM: Proceedings and Surveys  
Vortex-boundary element s i m ulation of three-dimensional wall-bounded ows is performed on a massively parallel architecture using the data parallel paradigm.  ...  In this paper, the CPU time and memory requirements for the direct evaluation of the vortical eld using three parallelization algorithms are compared.  ...  Andrei Malevsky who served as a consultant for questions regarding the CM5, and contributed directly towards the conversion of the code.  ... 
doi:10.1051/proc:1996011 fatcat:ephkjquzufdv5bsx26hwlfhdsa

Parallel Computation of Vortex Tube Reconnection Using a Graphics Card and the 3D Vortex–in–Cell Method

Henryk Kudela, Andrzej Kosior
2013 Procedia IUTAM  
We show that the VIC method is very well suited for parallel computation.  ...  For these numerical simulations we use the 3D vortex-in-cell (VIC) method.  ...  An important element of parallel computations is choosing the right numerical algorithm to allow for the effective use of computer architecture.  ... 
doi:10.1016/j.piutam.2013.03.008 fatcat:mq6quzrjojawbmox7e5gq4u6yu

Parallel computations on GPU in 3D using the vortex particle method

Andrzej Kosior, Henryk Kudela
2013 Computers & Fluids  
The paper presented the Vortex in Cell (VIC) method for solving the fluid motion equations in 3D and its implementation for parallel computation in multicore architecture of the Graphics Processing Unit  ...  Its speed-up was almost 12 times greater than for the direct Fast Poisson Solver for a single processor. The speed-up for the entire VIC method implementation on the GPU was 46 times.  ...  An important element of parallel computations is choosing the right computational method to allow for the effective use of computer architecture.  ... 
doi:10.1016/j.compfluid.2012.01.014 fatcat:xkug4xw2bzbc3ofrefri7gm7de

Billion vortex particle direct numerical simulations of aircraft wakes

Philippe Chatelain, Alessandro Curioni, Michael Bergdorf, Diego Rossinelli, Wanda Andreoni, Petros Koumoutsakos
2008 Computer Methods in Applied Mechanics and Engineering  
The simulations involve a highly efficient implementation of vortex methods on massively parallel computers, enabling unprecedented simulations using billions of particles.  ...  The remeshed particle locations are utilized for the computation of the field quantities, the discretization of the differential operators for diffusion and vortex stretching, and the solution of the Poisson  ...  Implementation for parallel computer architectures The proposed hybrid vortex particle methods encounter a number of challenges in their implementation in massively parallel architectures.  ... 
doi:10.1016/j.cma.2007.11.016 fatcat:dzozs5rll5gjhnantxmldxtrw4

The Implementation of the U.S. High Performance Computing and Communications Program [chapter]

Christian Bischof, Horst D. Simon
1992 Informatik aktuell  
We would also like to thank Thomas Morgan, Gail Pieper, Paul Plassmann, and Stephen Wright for their contributions to this summary report.  ...  This solution was computed using a new, inexact, damped Newton optimization algorithm developed by Paul Plassmann and Stephen Wright. .Figure 3 : 3 develop massively parallel architectures scalable to  ...  NASA is planning to .develop algorithm and architecture test beds capable of fully utilizing massively parallel concepts and increasing end-to-end performance, and .Earth and Space Sciences (ESS) -multi-disciplinary  ... 
doi:10.1007/978-3-642-77661-8_16 dblp:conf/supercomputer/BischofS92 fatcat:bnsel2voxrgu3bj3a46fu2fsiq

The 3D vortex particle method in parallel computations on many GPUs

Andrzej Kosior, Henryk Kudela
2014 Computers & Fluids  
Parallel implementation of the Vortex-in-Cell (VIC) method for 3D flow on many graphics cards was presented.  ...  Our aim was to show the great potential of the VIC method for solution of 3D flow problems and that it is very well suited for parallel computation.  ...  The VIC method is very well suited for parallel computation [11, 10, 19, 7] .  ... 
doi:10.1016/j.compfluid.2013.10.011 fatcat:olmsv65usvc3nilspsxqc7iry4

A Solver for Massively Parallel Direct Numerical Simulation of Three-Dimensional Multiphase Flows [article]

S. Shin, J. Chergui, D. Juric
2014 arXiv   pre-print
We present a new solver for massively parallel simulations of fully three-dimensional multiphase flows.  ...  The solver runs on a variety of computer architectures from laptops to supercomputers and on 65536 threads or more (limited only by the availability to us of more threads).  ...  ACKNOWLEDGEMENTS This work was supported by: (1) the Partnership for Advanced Computing in Europe (PRACE),  ... 
arXiv:1410.8568v1 fatcat:o5ear43lsvgndaux4e4tiiuhmu

Micromagnetic Simulations on GPU, A Case Study: Vortex Core Switching by High-Frequency Magnetic Fields

Ben Van de Wiele, Arne Vansteenkiste, Matthias Kammerer, Bartel Van Waeyenberge, Luc Dupre, Daniël De Zutter
2012 IEEE transactions on magnetics  
Since magnetic vortex cores have two ground states, they are candidates for digital memory bits in future magnetic random access memory (MRAM) devices.  ...  Vortex core switching can be induced by exciting the gyrotropic eigenmode, e.g., by applying cyclic magnetic fields with typically a sub-gigahertz frequency.  ...  Parallel computing has shown to be a very good solution for accelerating numerical software.  ... 
doi:10.1109/tmag.2011.2179551 fatcat:apvs3retv5bulm3x5mak2t3f74

MHD code using multi graphical processing units: SMAUG+

N. Gyenge, M.K. Griffiths, R. Erdélyi
2018 Advances in Space Research  
The main aim of the code development is to provide massively parallel code without the memory limitation of a single GPU. By using our code, the applied model size could be significantly increased.  ...  The initial configuration of the Orszag-Tang vortex simulations are distributed among 4, 16, 64 and 100 GPUs.  ...  for providing assistance with running the the SMAUG+ code on the Wilkes cluster.  ... 
doi:10.1016/j.asr.2017.10.027 fatcat:lquu6psznvfrhceth2ceq63qty

Experiences with the IBM SP1

W. D. Gropp, E. Lusk
1995 IBM Systems Journal  
It took only days, not months, to prepare for and migrate applications to this parallel supercomputer, demonstrating that high performance, parallelism, and portability can coexist.  ...  We explore what features of software technology and system architecture enabled immediate and successful use of the SP1.  ...  A promising approach for overcoming these limitations is the use of integral equation methods (IEMS) implemented on massively parallel computers.  ... 
doi:10.1147/sj.342.0249 fatcat:uidhyqvf5rey5mj3fzsakukq7u

Computation of incompressible flows with implicit finite element implementations on the Connection Machine

M. Behr, A. Johnson, J. Kennedy, S. Mittal, T. Tezduyar
1993 Computer Methods in Applied Mechanics and Engineering  
these methods on massively parallel architectures are discussed.  ...  Its current capability to solve large-scale problems, especially when coupled with the potential for growth enjoyed by massively parallel computers, make the implementation a worthwhile enterprise.  ...  Mathur of Thinking Machines Corporation for their advice and the CMSSL library routines used extensively in our finite element implementation. We would also  ... 
doi:10.1016/0045-7825(93)90155-q fatcat:djdvk77xcvfczlwazeazaut6gm

Rico and the jets: Direct numerical simulations of turbulent liquid jets

C. R. Constante-Amores, L. Kahouadji, A. Batchvarov, S. Shin, J. Chergui, D. Juric, O. K. Matar
2020 Physical Review Fluids  
We use a solver for massively parallel simulations of fully three-dimensional multiphase flows [10] , able to run on a variety of computer architectures, wholly written in FORTRAN 2008 and adopting an  ...  algebraic domain decomposition strategy for parallelization with MPI.  ... 
doi:10.1103/physrevfluids.5.110501 fatcat:uhmqvpjgxrfx5e2gesbt6e3zoy

Nodes, Modes and Flow Codes

George Em Karniadakis, Steven A. Orszag
1993 Physics today  
Thus the computational work requirement (embodied in the number of modes times the number of time steps) for DNS of turbulence scales roughly as R 3 and increases by an order of magnitude if R is doubled  ...  of freedom and simplify the computations.  ...  Equation 3, called Amdahl's law, shows that massively parallel computers with large P require massively parallelizable computations.  ... 
doi:10.1063/1.881374 fatcat:mvfkenopxvffnchc6s6t2tbzwq

Measuring and Comparing the Scaling Behaviour of a High-Performance CFD Code on Different Supercomputing Infrastructures

Jerome Frisch, Ralf-Peter Mundani
2015 2015 17th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC)  
Parallel code design is a challenging task especially when addressing petascale systems for massive parallel processing (MPP), i.e. parallel computations on several hundreds of thousands of cores.  ...  Basis for this code is an adaptive hierarchical data structure together with an efficient communication and (numerical) computation scheme that supports MPP.  ...  Furthermore, the authors would like to cordially thank Leibniz Supercomputing Centre (LRZ) in Garching for the computing time granted during the 'Extreme Scaling Workshop' in June 2014.  ... 
doi:10.1109/synasc.2015.63 dblp:conf/synasc/FrischM15 fatcat:ndu27dvg6nganb3v62icnt4ax4

High-Resolution Mesh Convergence Properties and Parallel Efficiency of a Spectral Element Atmospheric Dynamical Core

John Dennis, Aimé Fournier, William F. Spotz, Amik St-Cyr, Mark A. Taylor, Stephen J. Thomas, Henry Tufo
2005 The international journal of high performance computing applications  
Better than 80% parallel efficiency is obtained for fixed grids on up to 8938 processors. These  ...  We first demonstrate the parallel performance of the dynamical core of a spectral element atmospheric model.  ...  Parallel Scalability Spectral element methods are well suited to modern cache-based parallel computers for several reasons.  ... 
doi:10.1177/1094342005056108 fatcat:rdzo233qe5bu7pv5jibx4na5fm
« Previous Showing results 1 — 15 out of 2,870 results