Fast solvers for mesh-based computations
|
|
- Dulcie Kelly
- 5 years ago
- Views:
Transcription
1 Fast solvers for mesh-based computations Maciej Paszyński Department of Computer Science AGH University, Krakow, Poland home.agh.edu.pl/paszynsk Collaborators: Anna Paszyńska (Jagiellonian University,Poland) David Pardo (UPV/BCAM/IKERBASQUE,Spain) Victor Calo (UWA,Australia) Keshav Pingali (ICES,UT,USA) Luis Garcia-Castillo (Carlos III Univ. Madrid) Leszek Demkowicz (ICES,UT,USA) Mikhail Moshkov (KAUST,Saudi Arabia) Lisandro Dalcin (KAUST,Saudi Arabia) PhD Students: Piotr Gurgul (PhD def. 10/2014) Arkadiusz Szymczak (PhD def. 5/2015) Marcin Sieniek (PhD def. 11/2015) Maciej Woźniak Marcin Łoś Konrad Jopek Marcin Skotniczny Grzegorz Gurgul 1
2 Outline (global matrix storage, ordering, elimination trees, LU factorizations) PART I - Introduction 1. Sparse-matrix-based direct solvers 2. Mesh-based direct solvers (element partition trees, orderings, LU factorizations) PART II - Benefits of mesh-based solvers and element partition trees 1. New ordering algorithms 2. Transforming element partition tree into an ordering 3. Straightforward parallelization with element partition trees 4. Reutilization of partial LU factorizations 5. Reuse of identical LU factorizations 2
3 Computational mesh, sparse matrix and direct solvers Sparse matrices as seen by classical sparse-matrix-based direct solvers We miss some important information here: What is the structure of the mesh? Is it 1D, 2D or 3D? Uniform or h refined? What are the basis functions spread over the mesh? Are they uniform or p adaptive? What is the discretization method? 3
4 Computational mesh, sparse matrix and direct solvers Two dimensional mesh, finite element method, linear basis functions 4
5 Computational mesh, sparse matrix and direct solvers Two dimensional mesh, finite element method, quadratic basis functions 5
6 Computational mesh, sparse matrix and direct solvers Two dimensional mesh, isogeometric finite element method, quadratic B-splines 6
7 Computational mesh, sparse matrix and direct solvers Two dimensional mesh, isogeometric collocation method, quadratic B-splines 7
8 Sparse matrix based direct solvers Sparse global matrix, stored in some compressed manner, e.g. coordinate format, CSC format CSR format (see e.g. Sparse Matrix Computations lecture notes by Jean Yves L Excellent et al. for more details) 8
9 Sparse matrix based direct solvers Ordering generator Ordering P P -1 AP Ordering can be stored in a vector Several algorithms for constructing of the ordering looking at the structure of the sparse matrix, e.g. nested-dissections (METIS) aproximate minimum degree (AMD) PORD available through MUMPS solver interface 9
10 Sparse matrix based direct solvers Ordering generator Ordering P P -1 AP 1 X X X X X X X X X X 2 X X 0 X X X X 3 X X 0 X X X X 4 X X 0 X X X X 5 X X 0 X X X X 5 X X X X 4 X X 0 X X 3 X X 0 X X 2 X X 0 X X 1 X X X X X 0 X X X X Impact of ordering on factorization Elimination of the first row with two different orderings: 10
11 Sparse matrix based direct solvers Sparse-matrix-based solver Ordering generator Ordering P P -1 AP Elimination tree LU factorization Elimination tree is constructed internally by the solver The ordering defines elimination tree (is it 1 to 1?) Elimination tree expresses dependencies between particular steps of the factorization For more details on the elimination tree see e.g. Sparse Matrix Computations lecture notes by Jean Yves L Excellent et al..: It helps with memory management, parallelization of the factorizations) 11
12 Sparse matrix based direct solvers Sparse-matrix-based solver Ordering generator Ordering P P -1 AP Elimination tree LU factorization When you eliminate row a you need to subtract it from row c So row c is eliminated after row a When you eliminate row c or row f you need to subtract them from row g So row g is eliminated after rows c and f 12
13 Mesh-based solvers Element matrices Mesh-based solver LU factorization Element partition tree Element partition tree generator Mesh-based solver can perform LU factorization based on element partition tree list of element (dense) matrices 13
14 Mesh-based solvers LU factorization Ordering P P -1 AP Elimination tree Sparse-matrix-based solver Element partition tree Element partition tree generator Ordering for sparse-matrix-based solver can be also computed based on the element partition tree and pass it to sparse-matrix-based solver 14
15 Element partition trees Several algorithms has been proposed to generate element partition trees: Area & neighbors algorithm (2D grids h adaptive ) Paszyńska A., Paszyński M., Jopek K., Woźniak M., Goik D., Gurgul P., AbouEisha H., Moshkov M., Calo V. M., Lenharth V. M., Nguyen D., Pingali K., Quasi-Optimal Elimination Trees for 2D Grids with Singularities, Scientific Programming, Volume 2015 Article ID :1-18. Volume & neighbors algorith for (3D grids h adaptive ) Paszyńska A., Volume and neighbors algorithm for finding elimination trees for three dimensional h-adaptive grids, Computers & Mathematics with Applications, 68 (10) (2014) Bisections weighted by element size (2D, 3D, h and hp adaptive grids) H. AbouEisha, V. Calo, K. Jopek, M. Moshkov, A. Paszyńska, M. Paszyński, M.Skotniczny, Element Partition Trees for Two- and Three-Dimensional h-refined Meshes and Their Use to Optimize Direct Solver Performance, Journal of Computational Science (2016) submitted 15
16 Computational mesh, sparse matrix and direct solvers Bisection weighted by element size for 3D h adaptive and hp adaptive grids H. AbouEisha, V. Calo, K. Jopek, M. Moshkov, A. Paszyńska, M. Paszyński, M.Skotniczny, Element Partition Trees for Two- and Three-Dimensional h-refined Meshes and Their Use to Optimize Direct Solver Performance, Journal of Computational Science (2016) submitted 16
17 Element partition trees 17
18 Computational mesh, sparse matrix and direct solvers Bisections weighted by element size GRAPH: VERTICES = ELEMENTS WEIGHT = SCALLED ELEMENT SIZE * ORDER OF APPROXIMATION EDGES = ADJACENCY RELATION WEIGHT = ORDER OF APPROXIMATION 18
19 Computational mesh, sparse matrix and direct solvers Bisections weighted by element size 19
20 Element partition trees Ratio between Bisections weighted by element size and METIS: 1.52 Ratio between Bisections weighted by element size and AMD or PORD:
21 Ordering based on element partition tree 21
22 Ordering based on element partition tree Sparse-matrix-based solver Ordering generator Ordering P P -1 AP Elimination tree LU factorization Element partition tree can be translated into an ordering and passed to sparse matrix based solver 22
23 Ordering based on element partition tree Ordering is generated by post-order transition of the element partition tree, and listing nodes of the mesh that can be eliminated at this point 23
24 Ordering based on element partition tree List interior of element 1: 44 24
25 Ordering based on element partition tree List faces belonging to element 1 only: 44, 33,35,37,38,41 25
26 Ordering based on element partition tree List edges belonging to element 1 only: 44, 33,35,37,38,41, 16,17,23,24,13,28,20,27, 26
27 Ordering based on element partition tree List vertices belonging to element 1 only: 44, 33,35,37,38,41, 16,17,23,24,13,28,20,27, 1,2,8,7 27
28 Ordering based on element partition tree List interior of element 2: 44, 33,35,37,38,41, 16,17,23,24,13,28,20,27, 1,2,8,7, 45 28
29 Ordering based on element partition tree List faces belonging to element 2 only: 44, 33,35,37,38,41, 16,17,23,24,13,28,20,27, 1,2,8,7, 45, 34,36,39,40,43 29
30 Ordering based on element partition tree List edges belonging to element 2 only: 44, 33,35,37,38,41, 16,17,23,24,13,28,20,27, 1,2,8,7, 45, 34,36,39,40,43, 18,19,25,26,15,27,31,32 30
31 Ordering based on element partition tree List vertices belonging to element 2 only: 44, 33,35,37,38,41, 16,17,23,24,13,28,20,27, 1,2,8,7, 45, 34,36,39,40,43, 18,19,25,26,15,27,31,32, 5,6,12,11 31
32 Ordering based on element partition tree List faces belonging to elements 1 and 2: 44, 33,35,37,38,41, 16,17,23,24,13,28,20,27, 1,2,8,7, 45, 34,36,39,40,43, 18,19,25,26,15,27,31,32, 5,6,12,11, 42 32
33 Ordering based on element partition tree List faces belonging to elements 1 and 2: 44, 33,35,37,38,41, 16,17,23,24,13,28,20,27, 1,2,8,7, 45, 34,36,39,40,43, 18,19,25,26,15,27,31,32, 5,6,12,11, 42, 14,30,21,29 33
34 Ordering based on element partition tree List vertices belonging to elements 1 and 2: 44, 33,35,37,38,41, 16,17,23,24,13,28,20,27, 1,2,8,7, 45, 34,36,39,40,43, 18,19,25,26,15,27,31,32, 5,6,12,11, 42, 14,30,21,29, 3,4,10,9 34
35 Ordering based on element partition tree List vertices belonging to elements 1 and 2: 44, 33,35,37,38,41, 16,17,23,24,13,28,20,27, 1,2,8,7, 45, 34,36,39,40,43, 18,19,25,26,15,27,31,32, 5,6,12,11, 42, 14,30,21,29, 3,4,10,9 Send this ordering through PERM_IN into MUMPS solver 35
36 Ordering based on element partition tree The orderings generated by bisection weighted by element size are passed to MUMPS in PERM_IN array ( with option icntl(7)=1 ) ( MUMPS automatic icntl(7)=7 ) 36
37 Ordering based on element partition tree Internal nodes of element 1 Interface nades (between element 1 and 2) 37
38 Ordering based on element partition tree Internal nodes of element 2 Interface nades (between element 1 and 2) 38
39 Ordering based on element partition tree Interface nades (between element 1 and 2) 39
40 Straightforward parallelization Straighforward parallelization 44, 33,35,37,38,41, 16,17,23,24,13,28,20,27, 1,2,8,7, 45, 34,36,39,40,43, 18,19,25,26,15,27,31,32, 5,6,12,11, 42, 14,30,21,29, 3,4,10,9 Unfortunatelly this information cannot be send to MUMPS (it has to recover it by constructing its own elimination tree) 40
41 Straightforward parallelization Level by level processing of element partition trees Paszyńska A., Paszyński M., Jopek K., Woźniak M., Goik D., Gurgul P., AbouEisha H., Moshkov M., Calo V. M., Lenharth V. M., Nguyen D., Pingali K., Quasi-Optimal Elimination Trees for 2D Grids with Singularities, Scientific Programming, Volume 2015 Article ID :
42 Straightforward parallelization Level by level processing of element partition trees 42
43 Straightforward parallelization Level by level processing of element partition trees 43
44 Reutilization of partial LU factorizations Processing element partition tree Maciej Paszyński, David Pardo, Victor Calo, A direct solver with reutilization of LU factorizations for h-adaptive finite element grids with point singularities, Computers and Mathematics with Applications (2012) 65(8) Anna Paszyńska, Graph-grammar greedy algorithm for reutilization of partial LU factorization over 3D tetrahderal grids, Journal of Computational Science (2016) submitted 44
45 Reutilization of partial LU factorizations Processing element partition tree 45
46 Reutilization of partial LU factorizations Processing element partition tree 46
47 Reutilization of partial LU factorizations Processing element partition tree 47
48 Reutilization of partial LU factorizations Processing element partition tree 48
49 Reutilization of partial LU factorizations Processing element partition tree 49
50 Reutilization of partial LU factorizations Processing element partition tree 50
51 Reutilization of partial LU factorizations Updating the element partition tree when refining the mesh 51
52 Reutilization of partial LU factorizations Processing element partition tree 52
53 Reutilization of partial LU factorizations Updating the element partition tree when refining the mesh??? 53
54 Reutilization of partial LU factorizations Updating the element partition tree when refining the mesh optimistic case 54
55 Reutilization of partial LU factorizations Updating the element partition tree when refining the mesh pesimistic case 55
56 Reutilization of partial LU factorizations Updating the element partition tree when refining the mesh 56
57 Reuse of partial LU factorizations Marcin Sieniek, Maciej Paszyński, Subtree reuse in multi-frontal solvers on regular grids in Step-and-Flash Imprint Lithography Modeling, Advanced Engineering Materials (2014) 16(2) Ignacio Martinez-Fernandez, Maciej Woźniak, Luis-Garcia Castillo, Maciej Paszyński, Mesh-Based Multi-Frontal Solver with Reuse of Partial LU Factorizations for Antenna Array 57 Journal of Computational Science (2016) submitted
58 Reuse of partial LU factorizations Reuse of identical LU factors over identical branches of element partition tree 58
59 Reuse of partial LU factorizations Reuse of identical LU factors over identical branches of element partition tree 59
60 Reuse of partial LU factorizations First problem is a toy antenna approximated with 18 tetrahdrals 60
61 Reuse of partial LU factorizations Second problem is real antenna approximated with 3297 tetrahedrals 61
62 Reuse of partial LU factorizations Third problem is real antenna approximated with 8324 tetrahedrals 62
63 Conclusions Mesh-based solvers and element partition trees allow to speed up the LU factorization process This can be done by using 1. New ordering algorithms, such as Bisections weighted by element size 2. Straightforward parallelization based on the element partition trees 3. Reutilization of partial LU factorizations over unmodified mesh elements 4. Reuse of LU factorizations over identical parts of the mesh 63
64 Maciej Paszyński, Fast solvers for mesh-based computations, Taylor & Francis, CRC Press (Table of Contents + Introduction) 64
HYBRID DIRECT AND ITERATIVE SOLVER FOR H ADAPTIVE MESHES WITH POINT SINGULARITIES
HYBRID DIRECT AND ITERATIVE SOLVER FOR H ADAPTIVE MESHES WITH POINT SINGULARITIES Maciej Paszyński Department of Computer Science, AGH University of Science and Technology, Kraków, Poland email: paszynsk@agh.edu.pl
More informationOptimization of Execution Time, Energy Consumption, and Accuracy during Finite Element Method Simulations
Optimization of Execution Time, Energy Consumption, and Accuracy during Finite Element Method Simulations Maciej Woźniak, Marcin Łoś, Leszek Siwik, Darusz Król, Maciej Paszyński Department of Computer
More informationAPPLICATION OF PROJECTION-BASED INTERPOLATION ALGORITHM FOR NON-STATIONARY PROBLEM
Computer Science 7 (3) 26 http://dx.doi.org/.7494/csci.26.7.3.297 Maciej Woźniak Maciej Paszyński APPLICATION OF PROJECTION-BASED INTERPOLATION ALGORITHM FOR NON-STATIONARY PROBLEM Abstract In this paper,
More informationAlgorithms for construction of Element Partition Trees for Direct Solver executed over h refined grids with B-splines and C 0 separators
Available online at www.sciencedirect.com ScienceDirect Procedia Computer Science 108C (2017) 857 866 International Conference on Computational Science, ICCS 2017, 12-14 June 2017, Zurich, Switzerland
More informationA direct solver with reutilization of previously-computed LU factorizations for h-adaptive finite element grids with point singularities
Maciej Paszynski AGH University of Science and Technology Faculty of Computer Science, Electronics and Telecommunication Department of Computer Science al. A. Mickiewicza 30 30-059 Krakow, Poland tel.
More informationParallel multi-frontal solver for isogeometric finite element methods on GPU
Parallel multi-frontal solver for isogeometric finite element methods on GPU Maciej Paszyński Department of Computer Science, AGH University of Science and Technology, Kraków, Poland email: paszynsk@agh.edu.pl
More informationDynamic Programming Algorithm for Generation of Optimal Elimination Trees for Multi-Frontal Direct Solver over h-refined Grids
Procedia Computer Science Volume 29, 2014, Pages 947 959 ICCS 2014. 14th International Conference on Computational Science Dynamic Programming Algorithm for Generation of Optimal Elimination Trees for
More informationHypergraph Grammars in hp-adaptive Finite Element Method
Available online at www.sciencedirect.com Procedia Computer Science 18 (2013 ) 1545 1554 2013 International Conference on Computational Science Hypergraph Grammars in hp-adaptive Finite Element Method
More informationThe Development of a 3D Self-Adaptive Goal-Oriented hp-finite Element Software for Simulations of Resistivity Logging Instruments
A Progress Report on: The Development of a 3D Self-Adaptive Goal-Oriented hp-finite Element Software for Simulations of Resistivity Logging Instruments David Pardo (dzubiaur@yahoo.es), L. Demkowicz, C.
More informationNull space computation of sparse singular matrices with MUMPS
Null space computation of sparse singular matrices with MUMPS Xavier Vasseur (CERFACS) In collaboration with Patrick Amestoy (INPT-IRIT, University of Toulouse and ENSEEIHT), Serge Gratton (INPT-IRIT,
More informationNative mesh ordering with Scotch 4.0
Native mesh ordering with Scotch 4.0 François Pellegrini INRIA Futurs Project ScAlApplix pelegrin@labri.fr Abstract. Sparse matrix reordering is a key issue for the the efficient factorization of sparse
More informationOn the I/O Volume in Out-of-Core Multifrontal Methods with a Flexible Allocation Scheme
On the I/O Volume in Out-of-Core Multifrontal Methods with a Flexible Allocation Scheme Emmanuel Agullo 1,6,3,7, Abdou Guermouche 5,4,8, and Jean-Yves L Excellent 2,6,3,9 1 École Normale Supérieure de
More informationParallel hp-finite Element Simulations of 3D Resistivity Logging Instruments
Parallel hp-finite Element Simulations of 3D Resistivity Logging Instruments M. Paszyński 1,3, D. Pardo 1,2, L. Demkowicz 1, C. Torres-Verdin 2 1 Institute for Computational Engineering and Sciences 2
More informationAn algorithm for tensor product approximation of three-dimensional material data for implicit dynamics simulations
An algorithm for tensor product approximation of three-dimensional material data for implicit dynamics simulations Krzysztof Podsiadło, Marcin Łoś, Leszek Siwik, and Maciej Woźniak AGH University of Science
More informationA comparison of Algorithms for Sparse Matrix. Real-time Multibody Dynamic Simulation
A comparison of Algorithms for Sparse Matrix Factoring and Variable Reordering aimed at Real-time Multibody Dynamic Simulation Jose-Luis Torres-Moreno, Jose-Luis Blanco, Javier López-Martínez, Antonio
More informationOverview of MUMPS (A multifrontal Massively Parallel Solver)
Overview of MUMPS (A multifrontal Massively Parallel Solver) E. Agullo, P. Amestoy, A. Buttari, P. Combes, A. Guermouche, J.-Y. L Excellent, T. Slavova, B. Uçar CERFACS, ENSEEIHT-IRIT, INRIA (LIP and LaBRI),
More informationAMS526: Numerical Analysis I (Numerical Linear Algebra)
AMS526: Numerical Analysis I (Numerical Linear Algebra) Lecture 5: Sparse Linear Systems and Factorization Methods Xiangmin Jiao Stony Brook University Xiangmin Jiao Numerical Analysis I 1 / 18 Sparse
More informationParallel Greedy Matching Algorithms
Parallel Greedy Matching Algorithms Fredrik Manne Department of Informatics University of Bergen, Norway Rob Bisseling, University of Utrecht Md. Mostofa Patwary, University of Bergen 1 Outline Background
More informationF k G A S S1 3 S 2 S S V 2 V 3 V 1 P 01 P 11 P 10 P 00
PRLLEL SPRSE HOLESKY FTORIZTION J URGEN SHULZE University of Paderborn, Department of omputer Science Furstenallee, 332 Paderborn, Germany Sparse matrix factorization plays an important role in many numerical
More informationSCALABLE ALGORITHMS for solving large sparse linear systems of equations
SCALABLE ALGORITHMS for solving large sparse linear systems of equations CONTENTS Sparse direct solvers (multifrontal) Substructuring methods (hybrid solvers) Jacko Koster, Bergen Center for Computational
More informationLecture 4: Principles of Parallel Algorithm Design (part 4)
Lecture 4: Principles of Parallel Algorithm Design (part 4) 1 Mapping Technique for Load Balancing Minimize execution time Reduce overheads of execution Sources of overheads: Inter-process interaction
More informationContents. F10: Parallel Sparse Matrix Computations. Parallel algorithms for sparse systems Ax = b. Discretized domain a metal sheet
Contents 2 F10: Parallel Sparse Matrix Computations Figures mainly from Kumar et. al. Introduction to Parallel Computing, 1st ed Chap. 11 Bo Kågström et al (RG, EE, MR) 2011-05-10 Sparse matrices and storage
More informationCHARMS: A Simple Framework for Adaptive Simulation SIGGRAPH Presented by Jose Guerra
CHARMS: A Simple Framework for Adaptive Simulation SIGGRAPH 2002 Eitan Grinspun Caltech Petr Krysl UCSD Peter Schröder Caltech Presented by Jose Guerra 1 Outline Background Motivation (Element vs. Basis
More informationRoofline Model (Will be using this in HW2)
Parallel Architecture Announcements HW0 is due Friday night, thank you for those who have already submitted HW1 is due Wednesday night Today Computing operational intensity Dwarves and Motifs Stencil computation
More informationAMS526: Numerical Analysis I (Numerical Linear Algebra)
AMS526: Numerical Analysis I (Numerical Linear Algebra) Lecture 20: Sparse Linear Systems; Direct Methods vs. Iterative Methods Xiangmin Jiao SUNY Stony Brook Xiangmin Jiao Numerical Analysis I 1 / 26
More informationParallel Programming in the Age of Ubiquitous Parallelism. Keshav Pingali The University of Texas at Austin
Parallel Programming in the Age of Ubiquitous Parallelism Keshav Pingali The University of Texas at Austin Parallelism is everywhere Texas Advanced Computing Center Laptops Cell-phones Parallel programming?
More informationMemory Hierarchy Management for Iterative Graph Structures
Memory Hierarchy Management for Iterative Graph Structures Ibraheem Al-Furaih y Syracuse University Sanjay Ranka University of Florida Abstract The increasing gap in processor and memory speeds has forced
More informationSimulating tsunami propagation on parallel computers using a hybrid software framework
Simulating tsunami propagation on parallel computers using a hybrid software framework Xing Simula Research Laboratory, Norway Department of Informatics, University of Oslo March 12, 2007 Outline Intro
More informationEfficient Multigrid based solvers for Isogeometric Analysis
Efficient Multigrid based solvers for Isogeometric Analysis R. Tielen, M. Möller and C. Vuik Delft Institute of Applied Mathematics (DIAM) Numerical Analysis / 22 Isogeometric Analysis (IgA) Extension
More informationSparse Direct Factorizations through. Unassembled Hyper-Matrices
Sparse Direct Factorizations through Unassembled Hyper-Matrices Paolo Bientinesi, Victor Eijkhout, Kyungjoo Kim, Jason Kurtz, Robert van de Geijn August 27, 2009 keywordsfactorizations, Gaussian elimination,
More informationqr_mumps a multithreaded, multifrontal QR solver The MUMPS team,
qr_mumps a multithreaded, multifrontal QR solver The MUMPS team, MUMPS Users Group Meeting, May 29-30, 2013 The qr_mumps software the qr_mumps software qr_mumps version 1.0 a multithreaded, multifrontal
More informationMUMPS. The MUMPS library. Abdou Guermouche and MUMPS team, June 22-24, Univ. Bordeaux 1 and INRIA
The MUMPS library Abdou Guermouche and MUMPS team, Univ. Bordeaux 1 and INRIA June 22-24, 2010 MUMPS Outline MUMPS status Recently added features MUMPS and multicores? Memory issues GPU computing Future
More informationSparse Direct Factorizations through Unassembled Hyper-Matrices
TACC Technical Report TR-07-02 Sparse Direct Factorizations through Unassembled Hyper-Matrices Paolo Bientinesi, Victor Eijkhout, Kyungjoo Kim, Jason Kurtz, Robert van de Geijn November 3, 2009 This technical
More informationPetIGA. A Framework for High Performance Isogeometric Analysis. Santa Fe, Argentina. Knoxville, United States. Thuwal, Saudi Arabia
PetIGA A Framework for High Performance Isogeometric Analysis Lisandro Dalcin 1,3, Nathaniel Collier 2, Adriano Côrtes 3, Victor M. Calo 3 1 Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET)
More informationGraph drawing in spectral layout
Graph drawing in spectral layout Maureen Gallagher Colleen Tygh John Urschel Ludmil Zikatanov Beginning: July 8, 203; Today is: October 2, 203 Introduction Our research focuses on the use of spectral graph
More informationGraph and Hypergraph Partitioning for Parallel Computing
Graph and Hypergraph Partitioning for Parallel Computing Edmond Chow School of Computational Science and Engineering Georgia Institute of Technology June 29, 2016 Graph and hypergraph partitioning References:
More informationEffects of Solvers on Finite Element Analysis in COMSOL MULTIPHYSICS
Effects of Solvers on Finite Element Analysis in COMSOL MULTIPHYSICS Chethan Ravi B.R Dr. Venkateswaran P Corporate Technology - Research and Technology Center Siemens Technology and Services Private Limited
More informationAdvanced Computer Graphics
G22.2274 001, Fall 2010 Advanced Computer Graphics Project details and tools 1 Projects Details of each project are on the website under Projects Please review all the projects and come see me if you would
More informationSecond Conference on Parallel, Distributed, Grid and Cloud Computing for Engineering
State of the art distributed parallel computational techniques in industrial finite element analysis Second Conference on Parallel, Distributed, Grid and Cloud Computing for Engineering Ajaccio, France
More informationLong time integrations of a convective PDE on the sphere by RBF collocation
Long time integrations of a convective PDE on the sphere by RBF collocation Bengt Fornberg and Natasha Flyer University of Colorado NCAR Department of Applied Mathematics Institute for Mathematics Applied
More informationJournal of Engineering Research and Studies E-ISSN
Journal of Engineering Research and Studies E-ISS 0976-79 Research Article SPECTRAL SOLUTIO OF STEADY STATE CODUCTIO I ARBITRARY QUADRILATERAL DOMAIS Alavani Chitra R 1*, Joshi Pallavi A 1, S Pavitran
More informationLecture 4: Principles of Parallel Algorithm Design (part 3)
Lecture 4: Principles of Parallel Algorithm Design (part 3) 1 Exploratory Decomposition Decomposition according to a search of a state space of solutions Example: the 15-puzzle problem Determine any sequence
More informationADAPTIVE FINITE ELEMENT
Finite Element Methods In Linear Structural Mechanics Univ. Prof. Dr. Techn. G. MESCHKE SHORT PRESENTATION IN ADAPTIVE FINITE ELEMENT Abdullah ALSAHLY By Shorash MIRO Computational Engineering Ruhr Universität
More informationAdvanced Surface Based MoM Techniques for Packaging and Interconnect Analysis
Electrical Interconnect and Packaging Advanced Surface Based MoM Techniques for Packaging and Interconnect Analysis Jason Morsey Barry Rubin, Lijun Jiang, Lon Eisenberg, Alina Deutsch Introduction Fast
More informationSparse matrices: Basics
Outline : Basics Bora Uçar RO:MA, LIP, ENS Lyon, France CR-08: Combinatorial scientific computing, September 201 http://perso.ens-lyon.fr/bora.ucar/cr08/ 1/28 CR09 Outline Outline 1 Course presentation
More informationParallel resolution of sparse linear systems by mixing direct and iterative methods
Parallel resolution of sparse linear systems by mixing direct and iterative methods Phyleas Meeting, Bordeaux J. Gaidamour, P. Hénon, J. Roman, Y. Saad LaBRI and INRIA Bordeaux - Sud-Ouest (ScAlApplix
More informationSparse matrices, graphs, and tree elimination
Logistics Week 6: Friday, Oct 2 1. I will be out of town next Tuesday, October 6, and so will not have office hours on that day. I will be around on Monday, except during the SCAN seminar (1:25-2:15);
More informationStudy of Butterfly Patterns of Matrix in Interconnection Network
International Journal of Scientific & Engineering Research, Volume 7, Issue, December-6 3 ISSN 9-558 Study of Butterfly Patterns of Matrix in Interconnection Network Rakesh Kumar Katare Professor, Department
More informationSparse Matrices Direct methods
Sparse Matrices Direct methods Iain Duff STFC Rutherford Appleton Laboratory and CERFACS Summer School The 6th de Brùn Workshop. Linear Algebra and Matrix Theory: connections, applications and computations.
More informationBLAS and LAPACK + Data Formats for Sparse Matrices. Part of the lecture Wissenschaftliches Rechnen. Hilmar Wobker
BLAS and LAPACK + Data Formats for Sparse Matrices Part of the lecture Wissenschaftliches Rechnen Hilmar Wobker Institute of Applied Mathematics and Numerics, TU Dortmund email: hilmar.wobker@math.tu-dortmund.de
More informationJoint Advanced Student School 2007 Martin Dummer
Sierpiński-Curves Joint Advanced Student School 2007 Martin Dummer Statement of the Problem What is the best way to store a triangle mesh efficiently in memory? The following points are desired : Easy
More information3D Resistivity Logging Instruments
Parallel hp-finite Element Simulations of 3D Resistivity Logging Instruments M.Paszyński (1), D.Pardo (2), L.Demkowicz (3), C.Torres-Verdin (2) (1) Department of Modeling and Information Technology, AGH
More informationModel Reduction for High Dimensional Micro-FE Models
Model Reduction for High Dimensional Micro-FE Models Rudnyi E. B., Korvink J. G. University of Freiburg, IMTEK-Department of Microsystems Engineering, {rudnyi,korvink}@imtek.uni-freiburg.de van Rietbergen
More informationSparse Matrices Introduction to sparse matrices and direct methods
Sparse Matrices Introduction to sparse matrices and direct methods Iain Duff STFC Rutherford Appleton Laboratory and CERFACS Summer School The 6th de Brùn Workshop. Linear Algebra and Matrix Theory: connections,
More informationContents. I The Basic Framework for Stationary Problems 1
page v Preface xiii I The Basic Framework for Stationary Problems 1 1 Some model PDEs 3 1.1 Laplace s equation; elliptic BVPs... 3 1.1.1 Physical experiments modeled by Laplace s equation... 5 1.2 Other
More informationREDUCING GRAPH COLORING TO CLIQUE SEARCH
Asia Pacific Journal of Mathematics, Vol. 3, No. 1 (2016), 64-85 ISSN 2357-2205 REDUCING GRAPH COLORING TO CLIQUE SEARCH SÁNDOR SZABÓ AND BOGDÁN ZAVÁLNIJ Institute of Mathematics and Informatics, University
More informationPARDISO Version Reference Sheet Fortran
PARDISO Version 5.0.0 1 Reference Sheet Fortran CALL PARDISO(PT, MAXFCT, MNUM, MTYPE, PHASE, N, A, IA, JA, 1 PERM, NRHS, IPARM, MSGLVL, B, X, ERROR, DPARM) 1 Please note that this version differs significantly
More informationLecture 23: Starting to put it all together #2... More 2-Point Boundary value problems
Lecture 23: Starting to put it all together #2... More 2-Point Boundary value problems Outline 1) Our basic example again: -u'' + u = f(x); u(0)=α, u(l)=β 2) Solution of 2-point Boundary value problems
More informationModule 1: Introduction to Finite Difference Method and Fundamentals of CFD Lecture 6:
file:///d:/chitra/nptel_phase2/mechanical/cfd/lecture6/6_1.htm 1 of 1 6/20/2012 12:24 PM The Lecture deals with: ADI Method file:///d:/chitra/nptel_phase2/mechanical/cfd/lecture6/6_2.htm 1 of 2 6/20/2012
More informationA substructure based parallel dynamic solution of large systems on homogeneous PC clusters
CHALLENGE JOURNAL OF STRUCTURAL MECHANICS 1 (4) (2015) 156 160 A substructure based parallel dynamic solution of large systems on homogeneous PC clusters Semih Özmen, Tunç Bahçecioğlu, Özgür Kurç * Department
More informationParallel algorithms for Scientific Computing May 28, Hands-on and assignment solving numerical PDEs: experience with PETSc, deal.
Division of Scientific Computing Department of Information Technology Uppsala University Parallel algorithms for Scientific Computing May 28, 2013 Hands-on and assignment solving numerical PDEs: experience
More informationCost-Effective Parallel Computational Electromagnetic Modeling
Cost-Effective Parallel Computational Electromagnetic Modeling, Tom Cwik {Daniel.S.Katz, cwik}@jpl.nasa.gov Beowulf System at PL (Hyglac) l 16 Pentium Pro PCs, each with 2.5 Gbyte disk, 128 Mbyte memory,
More informationA parallel direct/iterative solver based on a Schur complement approach
A parallel direct/iterative solver based on a Schur complement approach Gene around the world at CERFACS Jérémie Gaidamour LaBRI and INRIA Bordeaux - Sud-Ouest (ScAlApplix project) February 29th, 2008
More information2D & 3D Finite Element Method Packages of CEMTool for Engineering PDE Problems
2D & 3D Finite Element Method Packages of CEMTool for Engineering PDE Problems Choon Ki Ahn, Jung Hun Park, and Wook Hyun Kwon 1 Abstract CEMTool is a command style design and analyzing package for scientific
More informationAutomated Aperture GenerationQuantitative Evaluation of Ape
. Objectives Approaches Randomness Reconstructing Images from K-space References Automated Aperture Generation Quantitative Evaluation of Aperture April 5, 011 . Objectives Approaches Randomness Reconstructing
More informationContents. Implementing the QR factorization The algebraic eigenvalue problem. Applied Linear Algebra in Geoscience Using MATLAB
Applied Linear Algebra in Geoscience Using MATLAB Contents Getting Started Creating Arrays Mathematical Operations with Arrays Using Script Files and Managing Data Two-Dimensional Plots Programming in
More informationAMS526: Numerical Analysis I (Numerical Linear Algebra)
AMS526: Numerical Analysis I (Numerical Linear Algebra) Lecture 1: Course Overview; Matrix Multiplication Xiangmin Jiao Stony Brook University Xiangmin Jiao Numerical Analysis I 1 / 21 Outline 1 Course
More informationChallenges in large-scale graph processing on HPC platforms and the Graph500 benchmark. by Nkemdirim Dockery
Challenges in large-scale graph processing on HPC platforms and the Graph500 benchmark by Nkemdirim Dockery High Performance Computing Workloads Core-memory sized Floating point intensive Well-structured
More informationEfficient Storage and Processing of Adaptive Triangular Grids using Sierpinski Curves
Efficient Storage and Processing of Adaptive Triangular Grids using Sierpinski Curves Csaba Attila Vigh, Dr. Michael Bader Department of Informatics, TU München JASS 2006, course 2: Numerical Simulation:
More informationA High Quality, Eulerian 3D Fluid Solver in C++ A Senior Project. presented to. the Faculty of the Computer Science Department of
A High Quality, Eulerian 3D Fluid Solver in C++ A Senior Project presented to the Faculty of the Computer Science Department of California Polytechnic State University, San Luis Obispo In Partial Fulfillment
More informationAn Edge-Swap Heuristic for Finding Dense Spanning Trees
Theory and Applications of Graphs Volume 3 Issue 1 Article 1 2016 An Edge-Swap Heuristic for Finding Dense Spanning Trees Mustafa Ozen Bogazici University, mustafa.ozen@boun.edu.tr Hua Wang Georgia Southern
More informationLecture 27: Fast Laplacian Solvers
Lecture 27: Fast Laplacian Solvers Scribed by Eric Lee, Eston Schweickart, Chengrun Yang November 21, 2017 1 How Fast Laplacian Solvers Work We want to solve Lx = b with L being a Laplacian matrix. Recall
More informationMemory Efficient Adaptive Mesh Generation and Implementation of Multigrid Algorithms Using Sierpinski Curves
Memory Efficient Adaptive Mesh Generation and Implementation of Multigrid Algorithms Using Sierpinski Curves Michael Bader TU München Stefanie Schraufstetter TU München Jörn Behrens AWI Bremerhaven Abstract
More informationIntroduction to Multigrid and its Parallelization
Introduction to Multigrid and its Parallelization! Thomas D. Economon Lecture 14a May 28, 2014 Announcements 2 HW 1 & 2 have been returned. Any questions? Final projects are due June 11, 5 pm. If you are
More informationPackage rmumps. December 18, 2017
Type Package Title Wrapper for MUMPS Library Version 5.1.2-2 Date 2017-12-13 Package rmumps December 18, 2017 Maintainer Serguei Sokol Description Some basic features of MUMPS
More informationOn Level Scheduling for Incomplete LU Factorization Preconditioners on Accelerators
On Level Scheduling for Incomplete LU Factorization Preconditioners on Accelerators Karl Rupp, Barry Smith rupp@mcs.anl.gov Mathematics and Computer Science Division Argonne National Laboratory FEMTEC
More informationA Moving Mesh Method for Time dependent Problems Based on Schwarz Waveform Relaxation
A Moving Mesh Method for Time dependent Problems Based on Schwarz Waveform Relaation Ronald D. Haynes, Weizhang Huang 2, and Robert D. Russell 3 Acadia University, Wolfville, N.S., Canada. ronald.haynes@acadiau.ca
More informationApproaches to Parallel Implementation of the BDDC Method
Approaches to Parallel Implementation of the BDDC Method Jakub Šístek Includes joint work with P. Burda, M. Čertíková, J. Mandel, J. Novotný, B. Sousedík. Institute of Mathematics of the AS CR, Prague
More informationParallel Hybrid Monte Carlo Algorithms for Matrix Computations
Parallel Hybrid Monte Carlo Algorithms for Matrix Computations V. Alexandrov 1, E. Atanassov 2, I. Dimov 2, S.Branford 1, A. Thandavan 1 and C. Weihrauch 1 1 Department of Computer Science, University
More informationLecture 15: More Iterative Ideas
Lecture 15: More Iterative Ideas David Bindel 15 Mar 2010 Logistics HW 2 due! Some notes on HW 2. Where we are / where we re going More iterative ideas. Intro to HW 3. More HW 2 notes See solution code!
More informationA Compiler for Parallel Finite Element Methods. with Domain-Decomposed Unstructured Meshes JONATHAN RICHARD SHEWCHUK AND OMAR GHATTAS
Contemporary Mathematics Volume 00, 0000 A Compiler for Parallel Finite Element Methods with Domain-Decomposed Unstructured Meshes JONATHAN RICHARD SHEWCHUK AND OMAR GHATTAS December 11, 1993 Abstract.
More informationFinite element discretization of Digital Material Representation models
Finite element discretization of Digital Material Representation models Lukasz Madej, Aleksander Fular, Krzysztof Banas, Filip Kruzel, Pawel Cybulka, Konrad Perzynski AGH University of Science and Technology,
More informationEvaluating the Performance of the Community Atmosphere Model at High Resolutions
Evaluating the Performance of the Community Atmosphere Model at High Resolutions Soumi Manna MS candidate, University of Wyoming Mentor: Dr. Ben Jamroz National Center for Atmospheric Research Boulder,
More informationA Moving Mesh Method for Time Dependent Problems based on Schwarz Waveform Relaxation
A Moving Mesh Method for Time Dependent Problems based on Schwarz Waveform Relaation Ronald D. Haynes, Weizhang Huang 2, and Robert D. Russell 3 Acadia University, Wolfville, N.S., Canada ronald.haynes@acadiau.ca
More informationA new 8-node quadrilateral spline finite element
Journal of Computational and Applied Mathematics 195 (2006) 54 65 www.elsevier.com/locate/cam A new 8-node quadrilateral spline finite element Chong-Jun Li, Ren-Hong Wang Institute of Mathematical Sciences,
More informationAdvanced Numerical Techniques for Cluster Computing
Advanced Numerical Techniques for Cluster Computing Presented by Piotr Luszczek http://icl.cs.utk.edu/iter-ref/ Presentation Outline Motivation hardware Dense matrix calculations Sparse direct solvers
More informationAdvanced Computer Graphics
G22.2274 001, Fall 2009 Advanced Computer Graphics Project details and tools 1 Project Topics Computer Animation Geometric Modeling Computational Photography Image processing 2 Optimization All projects
More informationMaths for Signals and Systems Linear Algebra in Engineering. Some problems by Gilbert Strang
Maths for Signals and Systems Linear Algebra in Engineering Some problems by Gilbert Strang Problems. Consider u, v, w to be non-zero vectors in R 7. These vectors span a vector space. What are the possible
More informationCoupled analysis of material flow and die deflection in direct aluminum extrusion
Coupled analysis of material flow and die deflection in direct aluminum extrusion W. Assaad and H.J.M.Geijselaers Materials innovation institute, The Netherlands w.assaad@m2i.nl Faculty of Engineering
More informationGraph Partitioning for High-Performance Scientific Simulations. Advanced Topics Spring 2008 Prof. Robert van Engelen
Graph Partitioning for High-Performance Scientific Simulations Advanced Topics Spring 2008 Prof. Robert van Engelen Overview Challenges for irregular meshes Modeling mesh-based computations as graphs Static
More informationLecture VIII. Global Approximation Methods: I
Lecture VIII Global Approximation Methods: I Gianluca Violante New York University Quantitative Macroeconomics G. Violante, Global Methods p. 1 /29 Global function approximation Global methods: function
More informationA Random Variable Shape Parameter Strategy for Radial Basis Function Approximation Methods
A Random Variable Shape Parameter Strategy for Radial Basis Function Approximation Methods Scott A. Sarra, Derek Sturgill Marshall University, Department of Mathematics, One John Marshall Drive, Huntington
More informationLecture 6: Input Compaction and Further Studies
PASI Summer School Advanced Algorithmic Techniques for GPUs Lecture 6: Input Compaction and Further Studies 1 Objective To learn the key techniques for compacting input data for reduced consumption of
More informationComputergrafik. Matthias Zwicker Universität Bern Herbst 2016
Computergrafik Matthias Zwicker Universität Bern Herbst 2016 Today Curves NURBS Surfaces Parametric surfaces Bilinear patch Bicubic Bézier patch Advanced surface modeling 2 Piecewise Bézier curves Each
More informationGRAPH CENTERS USED FOR STABILIZATION OF MATRIX FACTORIZATIONS
Discussiones Mathematicae Graph Theory 30 (2010 ) 349 359 GRAPH CENTERS USED FOR STABILIZATION OF MATRIX FACTORIZATIONS Pavla Kabelíková Department of Applied Mathematics FEI, VSB Technical University
More information13.472J/1.128J/2.158J/16.940J COMPUTATIONAL GEOMETRY
13.472J/1.128J/2.158J/16.940J COMPUTATIONAL GEOMETRY Lecture 23 Dr. W. Cho Prof. N. M. Patrikalakis Copyright c 2003 Massachusetts Institute of Technology Contents 23 F.E. and B.E. Meshing Algorithms 2
More informationSOL 6.13 Quadrilaterals
SOL 6.13 Quadrilaterals 6.13 The student will describe and identify properties of quadrilaterals. Understanding the Standard: A quadrilateral is a closed planar (two-dimensional) figure with four sides
More informationL21: Putting it together: Tree Search (Ch. 6)!
Administrative CUDA project due Wednesday, Nov. 28 L21: Putting it together: Tree Search (Ch. 6)! Poster dry run on Dec. 4, final presentations on Dec. 6 Optional final report (4-6 pages) due on Dec. 14
More informationSolving Sparse Linear Systems. Forward and backward substitution for solving lower or upper triangular systems
AMSC 6 /CMSC 76 Advanced Linear Numerical Analysis Fall 7 Direct Solution of Sparse Linear Systems and Eigenproblems Dianne P. O Leary c 7 Solving Sparse Linear Systems Assumed background: Gauss elimination
More informationLarge-scale Structural Analysis Using General Sparse Matrix Technique
Large-scale Structural Analysis Using General Sparse Matrix Technique Yuan-Sen Yang 1), Shang-Hsien Hsieh 1), Kuang-Wu Chou 1), and I-Chau Tsai 1) 1) Department of Civil Engineering, National Taiwan University,
More information