15 Hits in 10.8 sec

Towards Extreme-Scale Simulations with Next-Generation Trilinos: A Low Mach Fluid Application Case Study

Paul Lin, Matthew Bettencourt, Stefan Domino, Travis Fisher, Mark Hoemmen, Jonathan Hu, Eric Phipps, Andrey Prokopenko, Sivasankaran Rajamanickam, Christopher Siefert, Eric Cyr, Stephen Kennon
2014 2014 IEEE International Parallel & Distributed Processing Symposium Workshops  
We present a case study that involves integration of Trilinos with an engineering application (Sierra low Mach module/Nalu), involving the simulation of low Mach fluid flow for problems of size up to nine  ...  The next-generation Trilinos employs C++ templates in order to solve arbitrarily large problems and enable extreme-scale simulations.  ...  This paper describes the next-generation Trilinos and presents a case study of its integration into a new engineering application (SIERRA low Mach module/Nalu; henceforth referred to as "Nalu") that does  ... 
doi:10.1109/ipdpsw.2014.166 dblp:conf/ipps/LinBDFHHPPRSCK14 fatcat:oaakey2fcrbm7jwjlc3hah45se

Towards Extreme-Scale Simulations for Low Mach Fluids with Second-Generation Trilinos

Paul Lin, Matthew Bettencourt, Stefan Domino, Travis Fisher, Mark Hoemmen, Jonathan Hu, Eric Phipps, Andrey Prokopenko, Sivasankaran Rajamanickam, Christopher Siefert, Stephen Kennon
2014 Parallel Processing Letters  
We present a case study of the integration of Trilinos with a low Mach fluids engineering application (SIERRA low Mach module/Nalu).  ...  The second-generation Trilinos employs C++ templates in order to solve arbitrarily large problems.  ...  an application, as well as best practices. paper Towards Extreme-Scale Simulations with Second-Generation Trilinos 3 For perspective on the size of the simulations enabled by the new Nalu code compared  ... 
doi:10.1142/s0129626414420055 fatcat:q6ayjfvq35ftzgtthilxl4jnae

A survey of high level frameworks in block-structured adaptive mesh refinement packages

Anshu Dubey, Ann Almgren, John Bell, Martin Berzins, Steve Brandt, Greg Bryan, Phillip Colella, Daniel Graves, Michael Lijewski, Frank Löffler, Brian O'Shea, Erik Schnetter (+2 others)
2014 Journal of Parallel and Distributed Computing  
Some have stayed focused on specific domain areas, others have pursued a more general functionality, providing the building blocks for a larger variety of applications.  ...  The set consists of a mix of SAMR packages and application codes that cover a broad range of scientific domains.  ...  A case study of this kind of optimization can be found in [33] .  ... 
doi:10.1016/j.jpdc.2014.07.001 fatcat:uhzy4q53jfb3bjldz2tbl533jy

On the Memory Access Patterns of Supercomputer Applications: Benchmark Selection and Its Implications

Richard C. Murphy, Peter M. Kogge
2007 IEEE transactions on computers  
This paper compares the System Performance Evaluation Cooperative (SPEC) Integer and Floating-Point suites to a set of real-world applications for high-performance computing at Sandia National Laboratories  ...  These applications focus on the high-end scientific and engineering domains; however, the techniques presented in this paper are applicable to any application domain.  ...  The application analysis and data gathering discussed here used tools that were funded in part by the US Defense Advanced Research Projects Agency (DARPA) through Cray, Inc., under the High Productivity  ... 
doi:10.1109/tc.2007.1039 fatcat:jsmwvadsxfhhbhrcun5g2cggi4

A Monolithic Algebraic Multigrid Framework for Multiphysics Applications with Examples from Resistive MHD [article]

Peter Ohm, Tobias Wiesner, Eric C. Cyr, Jonathan J. Hu, John N. Shadid, Raymond S. Tuminaro
2021 arXiv   pre-print
A multigrid framework is described for multiphysics applications.  ...  Once defined, these components are ultimately assembled into a monolithic multigrid solver for the entire system.  ...  (a) t = 3.0s (b) t = 3.5s (c) t = 4.0s (d) t = 4.5s In this numerical study, we perform transient simulations of the problem with CFL max = 0.25 and CFL max = 0.5, and study the behavior of the nonlinear-and  ... 
arXiv:2103.07537v2 fatcat:z7o2t27q3vfbfcvsxzw24tbsai

A Jacobian-free Newton-Krylov method for time-implicit multidimensional hydrodynamics

M. Viallet, T. Goffrey, I. Baraffe, D. Folini, C. Geroux, M. V. Popov, J. Pratt, R. Walder
2016 Astronomy and Astrophysics  
Although our applications concern flows in stellar interiors, the method can be applied to general advection and/or diffusion-dominated flows.  ...  This work is a continuation of our efforts to develop an efficient implicit solver for multidimensional hydrodynamics for the purpose of studying important physical processes in stellar interiors, such  ...  M.V. acknowledges support from a Newton International Fellowship and Alumni program from the Royal Society during earlier part of this work.  ... 
doi:10.1051/0004-6361/201527339 fatcat:bwu25bxajzgalkh6h7mnlab7by

Scalable implicit incompressible resistive MHD with stabilized FE and fully-coupled Newton–Krylov-AMG

J.N. Shadid, R.P. Pawlowski, E.C. Cyr, R.S. Tuminaro, L. Chacón, P.D. Weber
2016 Computer Methods in Applied Mechanics and Engineering  
Initial results that explore the scaling of the solution methods are also presented on up to 128K processors for problems with up to 1.8B unknowns on a CrayXK7.  ...  These preconditioners are shown to enable a robust, scalable and efficient solution approach for the large-scale sparse linear systems generated by the Newton linearization.  ...  Mach number M A = u/u A = 3.  ... 
doi:10.1016/j.cma.2016.01.019 fatcat:fr632t2xvjbvnfsapu6tox2ev4

Towards a scalable fully-implicit fully-coupled resistive MHD formulation with stabilized FE methods

J.N. Shadid, R.P. Pawlowski, J.W. Banks, L. Chacón, P.T. Lin, R.S. Tuminaro
2010 Journal of Computational Physics  
Initial results that explore the scaling of the solution methods are presented on up to 4096 processors for problems with up to 64M unknowns on a CrayXT3/4.  ...  These preconditioners are shown to enable a robust, scalable and efficient solution approach for the large-scale sparse linear systems generated by the Newton linearization.  ...  This study is for a low Re = Re m = 0.7 flow with Ha = 1.  ... 
doi:10.1016/ fatcat:qubwtc354ba33k25bfqwayzdgq

Variational multiscale method for nonequilibrium plasma flows

Juan Pablo Trelles, S. Mahnaz Modirkhazeni
2014 Computer Methods in Applied Mechanics and Engineering  
The method is complemented with an intrinsic time-scales matrix definition to model the fluid-electromagnetic sub-grid scales inexpensively and a discontinuitycapturing operator to increase its robustness  ...  The resulting discrete system is solved by a generalized-alpha time-stepper together with a globalized inexact Newton-Krylov nonlinear solver.  ...  In the above limiting cases and for low speeds, the flow can be considered incompressible.  ... 
doi:10.1016/j.cma.2014.09.001 fatcat:hpce6cgfsjbnhh6zwrdequhjle

A staggered‐grid multilevel ILU for steady incompressible flows

Sven Baars, Mark van der Klok, Jonas Thies, Fred W. Wubs
2020 International Journal for Numerical Methods in Fluids  
Algorithms for studying transitions and instabilities in incompressible flows typically require the solution of linear systems with the full Jacobian matrix.  ...  In this paper, we identify some of the weak spots of the two-level scheme and propose remedies such as a different domain partitioning and recursive application of the method.  ...  Access to the SuperMUC-NG system during the 'friendly user phase' was kindly granted by LRZ Munich via a test account.  ... 
doi:10.1002/fld.4913 fatcat:oythgdajrbh3vizgm3evaad2ey

101 geodynamic modelling: how to design, interpret, and communicate numerical studies of the solid Earth

Iris van Zelst, Fabio Crameri, Adina E. Pusok, Anne Glerum, Juliane Dannberg, Cedric Thieulot
2022 Solid Earth  
Thus, with this perspective, we encourage high-quality modelling studies, fair external interpretation, and sensible use of published work.  ...  We clarify and consolidate terminology across geodynamics and numerical modelling to set a standard for clear communication of modelling studies.  ...  not commonly associated with fluids.  ... 
doi:10.5194/se-13-583-2022 fatcat:uingw7zsezhrbh3ejn4xi36ium

Parallel PDE-Based Simulations Using the Common Component Architecture [chapter]

Lois Curfman McInnes, Benjamin A. Allan, Robert Armstrong, Steven J. Benson, David E. Bernholdt, Tamara L. Dahlgren, Lori Freitag Diachin, Manojkumar Krishnan, James A. Kohl, J. Walter Larson, Sophia Lefantzi, Jarek Nieplocha (+4 others)
Lecture Notes in Computational Science and Engineering  
A goal of componentbased software engineering in such large-scale simulations is to help manage this complexity by enabling better interoperability among various codes that have been independently developed  ...  Complete parallel PDE-based applications can be built by combining reusable scientific components with application-specific components; both can employ core CCA services to manage inter-component interactions  ...  Some of the work in this paper was carried out by Northrop Grumman with funding provided by NASA's Computation Technologies (CT) Project, part of the Earth Science Technology Office (ESTO), under a contract  ... 
doi:10.1007/3-540-31619-1_10 fatcat:5poo3qaygbflxp2lejwym6usqu

Algorithms For Large Scale Problems In Eigenvalue And Svd Computations And In Big Data Applications

Lingfei Wu
Some applications often demand a fast solution of large, sparse eigenvalue and singular value problems; In other applications, extracting knowledge from large-scale data requires many techniques such as  ...  Our method may serve as a standalone kernel for providing a fast trace estimate or as a variance reduction method for Monte Carlo in some cases.  ...  In many cases, however, a low accuracy approximation is sufficient.  ... 
doi:10.21220/s2s880 fatcat:dji4if7if5gfre3c3q3dadpyw4

Hardware-aware solvers for large, sparse linear systems - Multi-precision and parallel approaches

Björn Henning Karl Rocker
The numerical simulation test case is a standard example in fluid dynamics: 3D stationary lid-driven cavity on a cuboid.  ...  Investigated is the potential of the SiCortex platform for numerical simulation by analyzing the performance of a set of elementary benchmarks and two fluid dynamics applications executed on the SC072  ... 
doi:10.5445/ir/1000023260 fatcat:td4xelwxsnaknfeeqx3irbg4hm

Portable shared-memory parallelisation strategies for high-order finite element codes

Jan Robert Eichstädt, Joaquim Peiró
Large-scale legacy solver frameworks thus face the question how to adapt their codebase in the most effective way, to enable performance and portability across a heterogeneous landscape of parallel HPC  ...  This thesis investigates strategies to parallelise numerical simulation software frameworks in the context of high-order finite element solvers.  ...  Direct numerical simulation (DNS) methods, that can accurately model the flow of any fluid continuum, are solving the full Navier-Stokes equations and spatially resolve the domain to extremely fine so  ... 
doi:10.25560/86621 fatcat:w4kfftnqavdcdg2j6bjhqk767i