Engineering Multilevel Graph Partitioning Algorithms

Size: px
Start display at page:

Download "Engineering Multilevel Graph Partitioning Algorithms"

Transcription

1 Engineering Multilevel Graph Partitioning Algorithms Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II 1 Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II KIT die Kooperation vonengineering Forschungszentrum Multilevel Graph Karlsruhe Partitioning GmbH Algorithms und Universität Karlsruhe (TH)

2 Overview Introduction Parallel Graph Partitioning (KaPPa) n-level Graph Partitioning (KaSPar) Experimental Evaluation Future Work 2 Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II

3 Simulation - FEM Simulation space is discretized into a mesh Solution of partitial differential equations are approximated by linear equations Number of vertices can become quite large time and memory Parallel processing required 3 Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II

4 The Common Parallel Approach Mesh partitioned via dual graph 1. Each volume (data, calculation) is represented by a vertex 2. Interdependencies are represented by edges All PE s get same amount of work Communication is expensive Graph Partitioning Problem: Partition a graph into (almost) equally sized blocks, such that the number of edges connecting vertices from different blocks is minimal. 4 Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II

5 ɛ-balanced Graph Partitioning Partition graph G = (V, E, c : V R >0, ω : E R >0 ) into k disjoint blocks s.t. total node weight of each block 1 + ɛ total node weight k total weight of cut edges as small as possible Applications: sparse matrix reordering, VLSI design, graph compression,... 5 Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II

6 Multi-Level Graph Partitioning Successful in existing systems: DiBaP, Chaco, Jostle, Metis, Scotch,... 6 Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II

7 Multi-Level Graph Partitioning Edge Contraction w w u, v x ω({x, z}) = ω({u, z}) + ω({v, z}). c(x) = c(u) + c(v) u v x 7 Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II

8 Why Yet Another MGP Algorithms? Karlsruhe Parallel Partitioner Karlsruhe Sequential Partitioner scalable parallel algorithm for the case k =# of processors large inputs high quality (better (even parallel) than other seq. algorithms) understand / engineer multi level method bridge gaps theory practice 8 Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II

9 Matching Selection Goals: 1. large edge weights sparsify 2. large #edges few levels 3. uniform node weights represent input 4. small node degrees represent input unclear objective gap to approx. weighted matching which only considers 1.,2. match contract input graph... Our Solution: Apply approx. weighted matching to general edge rating function 9 Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II

10 Edge Ratings Tried ω({u, v}) ω({u, v}) expansion({u, v}) := c(u) + c(v) expansion ω({u, v}) ({u, v}) := c(u)c(v) expansion 2 ω({u, v})2 ({u, v}) := c(u)c(v) ω({u, v}) innerouter({u, v}) := Out(v) + Out(u) 2ω(u, v) where c = node weight, ω =edge weight, Out(u) := {u,v} E ω({u, v}) match contract input graph Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II

11 Approx. Weighted Matching Use parallel prepartitioner (optional) currently geometric recursive bipartitioning if coords. available Global Path Algorithm [MaueSanders 2007] locally parallel [ManneBisseling 2007] for border nodes 11 Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II

12 Initial Partitioning Every PE can perform initial partitioning using different seeds Compared PARTY, PMETIS, and SCOTCH SCOTCH yielded best results input graph output partition local improvement match contract initial uncontract partitioning 12 Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II

13 Local Improvement Basic Approach Use linear time local search [Fiduccia Mattheyses 82] on multiple pairs of blocks in parallel input graph output partition... local improvement... match contract initial uncontract partitioning 13 Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II

14 Finding Pairs of Blocks Color edges of quotient graph Q several parallel algorithms tried Each color matching in Q independent block pairs 14 Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II

15 Pairwise Local Search exchange boundaries two local searches adopt best New queue selection strategy topgain 15 Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II

16 Karlsruhe Sequential Partitioner 1. Contraction contract a single edge between two levels possibly n levels finegrained contraction consequtive levels are very similar no matching algorithm required use edge rating expansion uniform distribution of node weights addressable priority queue: defines order of edges to be contracted dynamic graph data structure 2. Local Search efficent stopping criterion avoid quadratic runtime 3. General Trial Trees improve quality by independent trials 16 Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II

17 Local Search Nodes unmarked, active, marked active nodes u compute gains of moving from one block to another choose the maximum gain when moved active marked can t become active anymore unmarked neighbours of marked active v u v 17 Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II

18 Local Search - Stopping Criteria touching each node on each level leads to Ω(n 2 ) runtime need flexible stopping criteria gains in each step identically distributed, independent random variables expectation µ variance σ 2 compute µ and σ 2 from previous steps stop after p steps if pµ 2 > ασ 2 + β u v u v 18 Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II

19 Results Example Street network Europe V = 18M, E = 44M, k = 16, KaSPar ParMetis 19 Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II

20 Scalablity total time [s] KaPPa Strong KaPPa Fast KaPPa Minimal scotch kmetis parmetis k Street network Europe ( V = 18M, E = 44M) 20 Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II

21 Quality Comparison with Other Systems Geometric mean, imbalance ɛ = 0.03: 11 graphs (78K 18M nodes) k {2, 4, 8, 16, 64} algorithm large graphs best avg. t[s] KaFFPa strong KaSPar strong % KaPPa strong % Scotch % 3.83 kmetis % 0.71 Repeating scotch as long as KasPar strong run and choosing the best result 12.1% larger cuts Walshaw instances, road networks, Florida Sparse Matrix Collection, random Delaunay triangulations, random geometric graphs, social networks. 21 Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II

22 Sources of Quality Improvement 1. edge ratings largest effect by far 2. more localized refinement. Surprise! previous parallelizations caused worse quality 3. better queue selection 4. better matching algorithms 5. more time invested 22 Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II

23 Future Work KaFFPa 1. try other pairwise improvers (MaxFlow-MinCut) 2. try F-Cycles graph partitioning 3. better initial partitioning for large k 4. integration into Meta-heuristic 5. exploit shared memory parallelism δ l G s lb rb t b 1 b B 2 δr input graph match local improvement match... local improvement... contract initial uncontract contract uncontract partitioning 23 Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II

24 10th DIMACS Implementation Challenge Announcement Graph Partitioning and Clustering Organizers: 1. David Bader Georgia Institute of Technology 2. Peter Sanders and Dorothea Wagner, KIT 3. Web: 1st phase: call for instances and applications deadline soon 2nd phase: paper submission 3rd phase: challenge workshop 4th phase: proceedings Deadline for expressing interest in participation: 31. March 24 Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II

25 Thank you! Contact: 25 Mar. 3, 2011 Manuel Holtgrewe, Vitaly Osipov, Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II

Engineering Multilevel Graph Partitioning Algorithms

Engineering Multilevel Graph Partitioning Algorithms Engineering Multilevel Graph Partitioning Algorithms Peter Sanders, Christian Schulz Institute for Theoretical Computer Science, Algorithmics II 1 Nov. 10, 2011 Peter Sanders, Christian Schulz Institute

More information

Engineering State-of-the-Art Graph Partitioning

Engineering State-of-the-Art Graph Partitioning Engineering State-of-the-Art Graph Partitioning Libraries @KIT Vitaly Osipov, Peter Sanders, Christian Schulz, Manuel Holtgrewe Karlsruhe Institute of Technology, Karlsruhe, Germany Email: {osipov, sanders,

More information

Engineering Multilevel Graph Partitioning Algorithms

Engineering Multilevel Graph Partitioning Algorithms Engineering Multilevel Graph Partitioning Algorithms Peter Sanders, Christian Schulz Karlsruhe Institute of Technology (KIT), 76128 Karlsruhe, Germany {sanders,christian.schulz}@kit.edu Abstract. We present

More information

Think Locally, Act Globally: Highly Balanced Graph Partitioning

Think Locally, Act Globally: Highly Balanced Graph Partitioning Think Locally, Act Globally: Highly Balanced Graph Partitioning Peter Sanders, Christian Schulz Karlsruhe Institute of Technology, Karlsruhe, Germany {sanders, christian.schulz}@kit.edu Abstract. We present

More information

k-way Hypergraph Partitioning via n-level Recursive Bisection

k-way Hypergraph Partitioning via n-level Recursive Bisection k-way Hypergraph Partitioning via n-level Recursive Bisection Sebastian Schlag, Vitali Henne, Tobias Heuer, Henning Meyerhenke Peter Sanders, Christian Schulz January 10th, 2016 @ ALENEX 16 INSTITUTE OF

More information

Parallel Graph Partitioning for Complex Networks

Parallel Graph Partitioning for Complex Networks Parallel Graph Partitioning for Complex Networks Henning Meyerhenke, Peter Sanders, Christian Schulz High-performance Graph Algorithms and Applications in Computational Science Dagstuhl 1 Christian Schulz:

More information

Parallel Graph Partitioning for Complex Networks

Parallel Graph Partitioning for Complex Networks Parallel Graph Partitioning for Complex Networks Henning Meyerhenke Karlsruhe Institute of Technology (KIT) Karlsruhe, Germany meyerhenke@kit.edu Peter Sanders Karlsruhe Institute of Technology (KIT) Karlsruhe,

More information

Shape Optimizing Load Balancing for Parallel Adaptive Numerical Simulations Using MPI

Shape Optimizing Load Balancing for Parallel Adaptive Numerical Simulations Using MPI Parallel Adaptive Institute of Theoretical Informatics Karlsruhe Institute of Technology (KIT) 10th DIMACS Challenge Workshop, Feb 13-14, 2012, Atlanta 1 Load Balancing by Repartitioning Application: Large

More information

KaHIP v2.00 Karlsruhe High Quality Partitioning User Guide

KaHIP v2.00 Karlsruhe High Quality Partitioning User Guide KaHIP v2.00 Karlsruhe High Quality Partitioning User Guide Peter Sanders and Christian Schulz Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany Email: {sanders, christian.schulz}@kit.edu Abstract

More information

arxiv: v1 [cs.ne] 3 Oct 2011

arxiv: v1 [cs.ne] 3 Oct 2011 Distributed Evolutionary Graph Partitioning Peter Sanders, Christian Schulz Karlsruhe Institute of Technology, Karlsruhe, Germany Email: {sanders, christian.schulz}@kit.edu Abstract arxiv:1110.0477v1 [cs.ne]

More information

KaHIP v0.73 Karlsruhe High Quality Partitioning User Guide

KaHIP v0.73 Karlsruhe High Quality Partitioning User Guide KaHIP v0.73 Karlsruhe High Quality Partitioning User Guide Peter Sanders and Christian Schulz Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany Email: {sanders, christian.schulz}@kit.edu Abstract

More information

High-Quality Shared-Memory Graph Partitioning

High-Quality Shared-Memory Graph Partitioning High-Quality Shared-Memory Graph Partitioning Yaroslav Akhremtsev Karlsruhe Institute of Technology (KIT) yaroslav.akhremtsev@kit.edu Peter Sanders Karlsruhe Institute of Technology (KIT) peter.sanders@kit.edu

More information

Multi-Threaded Graph Partitioning

Multi-Threaded Graph Partitioning Multi-Threaded Graph Partitioning Dominique LaSalle and George Karypis Department of Computer Science & Engineering University of Minnesota Minneapolis, Minnesota 5555, USA {lasalle,karypis}@cs.umn.edu

More information

Graph and Hypergraph Partitioning for Parallel Computing

Graph 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 information

Graph Partitioning and Graph Clustering

Graph Partitioning and Graph Clustering 588 Graph Partitioning and Graph Clustering 10th DIMACS Implementation Challenge Workshop February 13 14, 2012 Georgia Institute of Technology Atlanta, GA David A. Bader Henning Meyerhenke Peter Sanders

More information

Advanced Coarsening Schemes for Graph Partitioning

Advanced Coarsening Schemes for Graph Partitioning Advanced Coarsening Schemes for Graph Partitioning ILYA SAFRO, Clemson University PETER SANDERS and CHRISTIAN SCHULZ, Karlsruhe Institute of Technology The graph partitioning problem is widely used and

More information

Shape Optimizing Load Balancing for MPI-Parallel Adaptive Numerical Simulations

Shape Optimizing Load Balancing for MPI-Parallel Adaptive Numerical Simulations Shape Optimizing Load Balancing for MPI-Parallel Adaptive Numerical Simulations Henning Meyerhenke Abstract. Load balancing is important for the efficient execution of numerical simulations on parallel

More information

Heuristic Graph Bisection with Less Restrictive Balance Constraints

Heuristic Graph Bisection with Less Restrictive Balance Constraints Heuristic Graph Bisection with Less Restrictive Balance Constraints Stefan Schamberger Fakultät für Elektrotechnik, Informatik und Mathematik Universität Paderborn Fürstenallee 11, D-33102 Paderborn schaum@uni-paderborn.de

More information

PuLP: Scalable Multi-Objective Multi-Constraint Partitioning for Small-World Networks

PuLP: Scalable Multi-Objective Multi-Constraint Partitioning for Small-World Networks PuLP: Scalable Multi-Objective Multi-Constraint Partitioning for Small-World Networks George M. Slota 1,2 Kamesh Madduri 2 Sivasankaran Rajamanickam 1 1 Sandia National Laboratories, 2 The Pennsylvania

More information

arxiv: v1 [cs.ne] 6 Feb 2017

arxiv: v1 [cs.ne] 6 Feb 2017 Distributed Evolutionary k-way Node Separators arxiv:1702.01692v1 [cs.ne] 6 Feb 2017 ABSTRACT Peter Sanders Karlsruhe Institute of Technology Karlsruhe, Germany sanders@kit.edu Darren Strash Colgate University

More information

Partitioning Complex Networks via Size-constrained Clustering

Partitioning Complex Networks via Size-constrained Clustering Partitioning Complex Networks via Size-constrained Clustering Henning Meyerhenke, Peter Sanders and Christian Schulz Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany Email: {meyerhenke, sanders,

More information

Engineering Edge Ratings and Matching Algorithms for Multilevel Graph Partitioning Algorithms

Engineering Edge Ratings and Matching Algorithms for Multilevel Graph Partitioning Algorithms Engineering Edge Ratings and Matching Algorithms for Multilevel Graph Partitioning Algorithms Maximilian Schuler Matriculation number: 1457790 December 7, 2011 Bachelor of Science in Computer Science at

More information

Think Locally, Act Globally: Perfectly Balanced Graph Partitioning

Think Locally, Act Globally: Perfectly Balanced Graph Partitioning Think Locally, ct Globally: Perfectly alanced Graph Partitioning Peter Sanders, hristian Schulz Karlsruhe Institute of Technology, Karlsruhe, Germany Email: {sanders, christian.schulz}@kit.edu arxiv:1210.0477v1

More information

Parallel repartitioning and remapping in

Parallel repartitioning and remapping in Parallel repartitioning and remapping in Sébastien Fourestier François Pellegrini November 21, 2012 Joint laboratory workshop Table of contents Parallel repartitioning Shared-memory parallel algorithms

More information

arxiv: v1 [cs.ds] 3 Apr 2017

arxiv: v1 [cs.ds] 3 Apr 2017 Graph Partitioning with Acyclicity Constraints Orlando Moreira 1, Merten Popp 1, and Christian Schulz 2 1 Intel Corporation, Eindhoven, The Netherlands, orlando.moreira@intel.com, merten.popp@intel.com

More information

Lecture 19: Graph Partitioning

Lecture 19: Graph Partitioning Lecture 19: Graph Partitioning David Bindel 3 Nov 2011 Logistics Please finish your project 2. Please start your project 3. Graph partitioning Given: Graph G = (V, E) Possibly weights (W V, W E ). Possibly

More information

PuLP. Complex Objective Partitioning of Small-World Networks Using Label Propagation. George M. Slota 1,2 Kamesh Madduri 2 Sivasankaran Rajamanickam 1

PuLP. Complex Objective Partitioning of Small-World Networks Using Label Propagation. George M. Slota 1,2 Kamesh Madduri 2 Sivasankaran Rajamanickam 1 PuLP Complex Objective Partitioning of Small-World Networks Using Label Propagation George M. Slota 1,2 Kamesh Madduri 2 Sivasankaran Rajamanickam 1 1 Sandia National Laboratories, 2 The Pennsylvania State

More information

arxiv: v2 [math.oc] 5 May 2015

arxiv: v2 [math.oc] 5 May 2015 Incorporating Road Networks into Territory Design Nitin Ahuja, Matthias Bender, Peter Sanders, Christian Schulz and Andreas Wagner a,b,c,c,a a PTV AG, {nitin.ahuja, andreas.wagner}@ptvgroup.com b FZI Research

More information

CS 140: Sparse Matrix-Vector Multiplication and Graph Partitioning

CS 140: Sparse Matrix-Vector Multiplication and Graph Partitioning CS 140: Sparse Matrix-Vector Multiplication and Graph Partitioning Parallel sparse matrix-vector product Lay out matrix and vectors by rows y(i) = sum(a(i,j)*x(j)) Only compute terms with A(i,j) 0 P0 P1

More information

Combining Recursive Bisection and k-way Local Search for Hypergraph Partitioning Charel Mercatoris

Combining Recursive Bisection and k-way Local Search for Hypergraph Partitioning Charel Mercatoris Bachelor Thesis Combining Recursive Bisection and k-way Local Search for Hypergraph Partitioning Charel Mercatoris Date: 8. November 2018 Supervisors: Prof. Dr. rer. nat. Peter Sanders Sebastian Schlag,

More information

Efficient Nested Dissection for Multicore Architectures

Efficient Nested Dissection for Multicore Architectures Efficient Nested Dissection for Multicore Architectures Dominique Lasalle and George Karypis Department of Computer Science & Engineering, University of Minnesota, Minneapolis, MN 55455, USA {lasalle,karypis}@cs.umn.edu

More information

A Divisive clustering technique for maximizing the modularity

A Divisive clustering technique for maximizing the modularity A Divisive clustering technique for maximizing the modularity Ümit V. Çatalyürek, Kamer Kaya, Johannes Langguth, and Bora Uçar February 13, 2012 1 Introduction 2 Clustering Paradigms 3 Algorithm 4 Results

More information

Overview of Trilinos and PT-Scotch

Overview of Trilinos and PT-Scotch 29.03.2012 Outline PT-Scotch 1 PT-Scotch The Dual Recursive Bipartitioning Algorithm Parallel Graph Bipartitioning Methods 2 Overview of the Trilinos Packages Examples on using Trilinos PT-Scotch The Scotch

More information

Parallel Algorithm for Multilevel Graph Partitioning and Sparse Matrix Ordering

Parallel Algorithm for Multilevel Graph Partitioning and Sparse Matrix Ordering Parallel Algorithm for Multilevel Graph Partitioning and Sparse Matrix Ordering George Karypis and Vipin Kumar Brian Shi CSci 8314 03/09/2017 Outline Introduction Graph Partitioning Problem Multilevel

More information

New Challenges In Dynamic Load Balancing

New Challenges In Dynamic Load Balancing New Challenges In Dynamic Load Balancing Karen D. Devine, et al. Presentation by Nam Ma & J. Anthony Toghia What is load balancing? Assignment of work to processors Goal: maximize parallel performance

More information

Requirements of Load Balancing Algorithm

Requirements of Load Balancing Algorithm LOAD BALANCING Programs and algorithms as graphs Geometric Partitioning Graph Partitioning Recursive Graph Bisection partitioning Recursive Spectral Bisection Multilevel Graph partitioning Hypergraph Partitioning

More information

arxiv: v1 [cs.ds] 20 Feb 2018

arxiv: v1 [cs.ds] 20 Feb 2018 Alexandra Henzinger 1, Alexander Noe 2, and Christian Schulz 3 1 Stanford University, Stanford, CA, USA ahenz@stanford.edu 2 University of Vienna, Vienna, Austria alexander.noe@univie.ac.at 3 University

More information

CHAPTER 6 DEVELOPMENT OF PARTICLE SWARM OPTIMIZATION BASED ALGORITHM FOR GRAPH PARTITIONING

CHAPTER 6 DEVELOPMENT OF PARTICLE SWARM OPTIMIZATION BASED ALGORITHM FOR GRAPH PARTITIONING CHAPTER 6 DEVELOPMENT OF PARTICLE SWARM OPTIMIZATION BASED ALGORITHM FOR GRAPH PARTITIONING 6.1 Introduction From the review, it is studied that the min cut k partitioning problem is a fundamental partitioning

More information

Graph Partitioning with Natural Cuts

Graph Partitioning with Natural Cuts Graph Partitioning with Natural Cuts Daniel Delling Andrew V. Goldberg Ilya Razenshteyn 1 Renato F. Werneck Microsoft Research Silicon Valley, 1065 La Avenida, Mountain View, CA 94043, USA December 2010

More information

Graph 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 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 information

Native mesh ordering with Scotch 4.0

Native 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 information

A Multi-Level Framework for Bisection Heuristics

A Multi-Level Framework for Bisection Heuristics A Multi-Level Framework for Bisection Heuristics Student thesis at the Institut for Theoretical Computer Science, Algorithmic I Prof. Dr. Dorothea Wagner Faculty of Informatics Karlsruhe Institute of Technology

More information

Parallel Mesh Partitioning in Alya

Parallel Mesh Partitioning in Alya Available online at www.prace-ri.eu Partnership for Advanced Computing in Europe Parallel Mesh Partitioning in Alya A. Artigues a *** and G. Houzeaux a* a Barcelona Supercomputing Center ***antoni.artigues@bsc.es

More information

Multilevel Algorithms for Multi-Constraint Hypergraph Partitioning

Multilevel Algorithms for Multi-Constraint Hypergraph Partitioning Multilevel Algorithms for Multi-Constraint Hypergraph Partitioning George Karypis University of Minnesota, Department of Computer Science / Army HPC Research Center Minneapolis, MN 55455 Technical Report

More information

Abusing a hypergraph partitioner for unweighted graph partitioning

Abusing a hypergraph partitioner for unweighted graph partitioning Abusing a hypergraph partitioner for unweighted graph partitioning B. O. Fagginger Auer R. H. Bisseling Utrecht University February 13, 2012 Fagginger Auer, Bisseling (UU) Mondriaan Graph Partitioning

More information

On Partitioning FEM Graphs using Diffusion

On Partitioning FEM Graphs using Diffusion On Partitioning FEM Graphs using Diffusion Stefan Schamberger Universität Paderborn, Fakultät für Elektrotechnik, Informatik und Mathematik Fürstenallee 11, D-33102 Paderborn email: schaum@uni-paderborn.de

More information

Improved Algebraic Cryptanalysis of QUAD, Bivium and Trivium via Graph Partitioning on Equation Systems

Improved Algebraic Cryptanalysis of QUAD, Bivium and Trivium via Graph Partitioning on Equation Systems Improved Algebraic of QUAD, Bivium and Trivium via on Equation Systems Kenneth Koon-Ho Wong 1, Gregory V. Bard 2 1 Information Security Institute Queensland University of Technology, Brisbane, Australia

More information

Multilevel Graph Partitioning

Multilevel Graph Partitioning Multilevel Graph Partitioning George Karypis and Vipin Kumar Adapted from Jmes Demmel s slide (UC-Berkely 2009) and Wasim Mohiuddin (2011) Cover image from: Wang, Wanyi, et al. "Polygonal Clustering Analysis

More information

PULP: Fast and Simple Complex Network Partitioning

PULP: Fast and Simple Complex Network Partitioning PULP: Fast and Simple Complex Network Partitioning George Slota #,* Kamesh Madduri # Siva Rajamanickam * # The Pennsylvania State University *Sandia National Laboratories Dagstuhl Seminar 14461 November

More information

Algorithm XXX: Mongoose, A Graph Coarsening and Partitioning Library

Algorithm XXX: Mongoose, A Graph Coarsening and Partitioning Library 0 Algorithm XXX: Mongoose, A Graph Coarsening and Partitioning Library TIMOTHY A. DAVIS, Texas A&M University WILLIAM W. HAGER, University of Florida SCOTT P. KOLODZIEJ, Texas A&M University S. NURI YERALAN,

More information

Evolutionary k-way Node Separators

Evolutionary k-way Node Separators Bachelor thesis Evolutionary k-way Node Separators Robert Williger Date: 6. November 2016 Supervisors: Prof. Dr. Peter Sanders Dr. Christian Schulz Dr. Darren Strash Institute of Theoretical Informatics,

More information

Parallel Graph Partitioning on a CPU-GPU Architecture

Parallel Graph Partitioning on a CPU-GPU Architecture Parallel Graph Partitioning on a CPU-GPU Architecture Bahareh Goodarzi Martin Burtscher Dhrubajyoti Goswami Department of Computer Science Department of Computer Science Department of Computer Science

More information

Highway Hierarchies Star

Highway Hierarchies Star Delling/Sanders/Schultes/Wagner: Highway Hierarchies Star 1 Highway Hierarchies Star Daniel Delling Dominik Schultes Peter Sanders Dorothea Wagner Institut für Theoretische Informatik Algorithmik I/II

More information

Application of Fusion-Fission to the multi-way graph partitioning problem

Application of Fusion-Fission to the multi-way graph partitioning problem Application of Fusion-Fission to the multi-way graph partitioning problem Charles-Edmond Bichot Laboratoire d Optimisation Globale, École Nationale de l Aviation Civile/Direction des Services de la Navigation

More information

Optimizing Irregular Adaptive Applications on Multi-threaded Processors: The Case of Medium-Grain Parallel Delaunay Mesh Generation

Optimizing Irregular Adaptive Applications on Multi-threaded Processors: The Case of Medium-Grain Parallel Delaunay Mesh Generation Optimizing Irregular Adaptive Applications on Multi-threaded rocessors: The Case of Medium-Grain arallel Delaunay Mesh Generation Filip Blagojević The College of William & Mary CSci 710 Master s roject

More information

Exact Combinatorial Algorithms for Graph Bisection

Exact Combinatorial Algorithms for Graph Bisection Exact Combinatorial Algorithms for Graph Bisection Daniel Delling Andrew V. Goldberg Ilya Razenshteyn Renato Werneck Microsoft Research Silicon Valley DIMACS Challenge Renato Werneck (Microsoft Research)

More information

Improving Graph Partitioning for Modern Graphs and Architectures

Improving Graph Partitioning for Modern Graphs and Architectures Improving Graph Partitioning for Modern Graphs and Architectures Dominique LaSalle lasalle@cs.umn.edu Narayanan Sundaram * narayanan.sundaram@intel.com Md Mostofa Ali Patwary * mostofa.ali.patwary@intel.com

More information

Efficient Parallel and External Matching

Efficient Parallel and External Matching Efficient Parallel and External Matching Marcel Birn, Vitaly Osipov, Peter Sanders, Christian Schulz, Nodari Sitchinava Karlsruhe Institute of Technology, Karlsruhe, Germany marcelbirn@gmx.de,{osipov,sanders,christian.schulz}@kit.edu,nodari@ira.uka.de

More information

Meshlization of Irregular Grid Resource Topologies by Heuristic Square-Packing Methods

Meshlization of Irregular Grid Resource Topologies by Heuristic Square-Packing Methods Meshlization of Irregular Grid Resource Topologies by Heuristic Square-Packing Methods Uei-Ren Chen 1, Chin-Chi Wu 2, and Woei Lin 3 1 Department of Electronic Engineering, Hsiuping Institute of Technology

More information

Parallel FEM Computation and Multilevel Graph Partitioning Xing Cai

Parallel FEM Computation and Multilevel Graph Partitioning Xing Cai Parallel FEM Computation and Multilevel Graph Partitioning Xing Cai Simula Research Laboratory Overview Parallel FEM computation how? Graph partitioning why? The multilevel approach to GP A numerical example

More information

Hypergraph-Partitioning Based Decomposition for Parallel Sparse-Matrix Vector Multiplication

Hypergraph-Partitioning Based Decomposition for Parallel Sparse-Matrix Vector Multiplication Hypergraph-Partitioning Based Decomposition for Parallel Sparse-Matrix Vector Multiplication Ümit V. Çatalyürek and Cevdet Aykanat, Member, IEEE Computer Engineering Department, Bilkent University 06 Bilkent,

More information

arxiv: v1 [cs.ds] 14 Feb 2017

arxiv: v1 [cs.ds] 14 Feb 2017 Optimal Longest Paths by Dynamic Programming Tomáš Balyo 1, Kai Fieger 1, and Christian Schulz 1,2 1 Karlsruhe Institute of Technology, Karlsruhe, Germany {tomas.balyo, christian.schulz}@kit.edu, fieger@ira.uka.de

More information

CS 473: Algorithms. Ruta Mehta. Spring University of Illinois, Urbana-Champaign. Ruta (UIUC) CS473 1 Spring / 36

CS 473: Algorithms. Ruta Mehta. Spring University of Illinois, Urbana-Champaign. Ruta (UIUC) CS473 1 Spring / 36 CS 473: Algorithms Ruta Mehta University of Illinois, Urbana-Champaign Spring 2018 Ruta (UIUC) CS473 1 Spring 2018 1 / 36 CS 473: Algorithms, Spring 2018 LP Duality Lecture 20 April 3, 2018 Some of the

More information

Can Recursive Bisection Alone Produce Routable Placements?

Can Recursive Bisection Alone Produce Routable Placements? Supported by Cadence Can Recursive Bisection Alone Produce Routable Placements? Andrew E. Caldwell Andrew B. Kahng Igor L. Markov http://vlsicad.cs.ucla.edu Outline l Routability and the placement context

More information

Shared memory parallel algorithms in Scotch 6

Shared memory parallel algorithms in Scotch 6 Shared memory parallel algorithms in Scotch 6 François Pellegrini EQUIPE PROJET BACCHUS Bordeaux Sud-Ouest 29/05/2012 Outline of the talk Context Why shared-memory parallelism in Scotch? How to implement

More information

XtraPuLP. Partitioning Trillion-edge Graphs in Minutes. State University

XtraPuLP. Partitioning Trillion-edge Graphs in Minutes. State University XtraPuLP Partitioning Trillion-edge Graphs in Minutes George M. Slota 1 Sivasankaran Rajamanickam 2 Kamesh Madduri 3 Karen Devine 2 1 Rensselaer Polytechnic Institute, 2 Sandia National Labs, 3 The Pennsylvania

More information

Parallel Graph Partitioning on Multicore Architectures

Parallel Graph Partitioning on Multicore Architectures Parallel Graph Partitioning on Multicore Architectures Xin Sui 1, Donald Nguyen 1, Martin Burtscher 2, and Keshav Pingali 1,3 1 Department of Computer Science, University of Texas at Austin 2 Department

More information

Order or Shuffle: Empirically Evaluating Vertex Order Impact on Parallel Graph Computations

Order or Shuffle: Empirically Evaluating Vertex Order Impact on Parallel Graph Computations Order or Shuffle: Empirically Evaluating Vertex Order Impact on Parallel Graph Computations George M. Slota 1 Sivasankaran Rajamanickam 2 Kamesh Madduri 3 1 Rensselaer Polytechnic Institute, 2 Sandia National

More information

Some aspects of parallel program design. R. Bader (LRZ) G. Hager (RRZE)

Some aspects of parallel program design. R. Bader (LRZ) G. Hager (RRZE) Some aspects of parallel program design R. Bader (LRZ) G. Hager (RRZE) Finding exploitable concurrency Problem analysis 1. Decompose into subproblems perhaps even hierarchy of subproblems that can simultaneously

More information

Evolutionary Multi-Level Acyclic Graph Partitioning

Evolutionary Multi-Level Acyclic Graph Partitioning lutionary Acyclic Graph Partitioning Orlando Moreira Intel Corporation Eindhoven, The Netherlands orlando.moreira@intel.com Merten Popp Intel Corporation Eindhoven, The Netherlands merten.popp@intel.com

More information

Parallel Graph Partitioning on Multicore Architectures

Parallel Graph Partitioning on Multicore Architectures Parallel Graph Partitioning on Multicore Architectures Xin Sui 1, Donald Nguyen 1, and Keshav Pingali 1,2 1 Department of Computer Science, University of Texas, Austin 2 Institute for Computational Engineering

More information

Penalized Graph Partitioning for Static and Dynamic Load Balancing

Penalized Graph Partitioning for Static and Dynamic Load Balancing Penalized Graph Partitioning for Static and Dynamic Load Balancing Tim Kiefer, Dirk Habich, Wolfgang Lehner Euro-Par 06, Grenoble, France, 06-08-5 Task Allocation Challenge Application (Workload) = Set

More information

Parallel Multilevel Algorithms for Multi-constraint Graph Partitioning

Parallel Multilevel Algorithms for Multi-constraint Graph Partitioning Parallel Multilevel Algorithms for Multi-constraint Graph Partitioning Kirk Schloegel, George Karypis, and Vipin Kumar Army HPC Research Center Department of Computer Science and Engineering University

More information

High Level Synthesis

High Level Synthesis High Level Synthesis Design Representation Intermediate representation essential for efficient processing. Input HDL behavioral descriptions translated into some canonical intermediate representation.

More information

Partitioning and Partitioning Tools. Tim Barth NASA Ames Research Center Moffett Field, California USA

Partitioning and Partitioning Tools. Tim Barth NASA Ames Research Center Moffett Field, California USA Partitioning and Partitioning Tools Tim Barth NASA Ames Research Center Moffett Field, California 94035-00 USA 1 Graph/Mesh Partitioning Why do it? The graph bisection problem What are the standard heuristic

More information

Performance Optimization of Parallel Lattice Boltzmann Fluid Flow Simulations on Complex Geometries

Performance Optimization of Parallel Lattice Boltzmann Fluid Flow Simulations on Complex Geometries Institute for Applied and Numerical Mathematics 4 Prof. Dr. Vincent Heuveline Performance Optimization of Parallel Lattice Boltzmann Fluid Flow Simulations on Complex Geometries Diploma Thesis of Jonas

More information

Jure Leskovec, Cornell/Stanford University. Joint work with Kevin Lang, Anirban Dasgupta and Michael Mahoney, Yahoo! Research

Jure Leskovec, Cornell/Stanford University. Joint work with Kevin Lang, Anirban Dasgupta and Michael Mahoney, Yahoo! Research Jure Leskovec, Cornell/Stanford University Joint work with Kevin Lang, Anirban Dasgupta and Michael Mahoney, Yahoo! Research Network: an interaction graph: Nodes represent entities Edges represent interaction

More information

Big Data Analytics. Special Topics for Computer Science CSE CSE Feb 11

Big Data Analytics. Special Topics for Computer Science CSE CSE Feb 11 Big Data Analytics Special Topics for Computer Science CSE 4095-001 CSE 5095-005 Feb 11 Fei Wang Associate Professor Department of Computer Science and Engineering fei_wang@uconn.edu Clustering II Spectral

More information

PARALLEL DECOMPOSITION OF 100-MILLION DOF MESHES INTO HIERARCHICAL SUBDOMAINS

PARALLEL DECOMPOSITION OF 100-MILLION DOF MESHES INTO HIERARCHICAL SUBDOMAINS Technical Report of ADVENTURE Project ADV-99-1 (1999) PARALLEL DECOMPOSITION OF 100-MILLION DOF MESHES INTO HIERARCHICAL SUBDOMAINS Hiroyuki TAKUBO and Shinobu YOSHIMURA School of Engineering University

More information

CAD Algorithms. Circuit Partitioning

CAD Algorithms. Circuit Partitioning CAD Algorithms Partitioning Mohammad Tehranipoor ECE Department 13 October 2008 1 Circuit Partitioning Partitioning: The process of decomposing a circuit/system into smaller subcircuits/subsystems, which

More information

MULTILEVEL OPTIMIZATION OF GRAPH BISECTION WITH PHEROMONES

MULTILEVEL OPTIMIZATION OF GRAPH BISECTION WITH PHEROMONES MULTILEVEL OPTIMIZATION OF GRAPH BISECTION WITH PHEROMONES Peter Korošec Computer Systems Department Jožef Stefan Institute, Ljubljana, Slovenia peter.korosec@ijs.si Jurij Šilc Computer Systems Department

More information

Jessica Su (some parts copied from CLRS / last quarter s notes)

Jessica Su (some parts copied from CLRS / last quarter s notes) 1 Max flow Consider a directed graph G with positive edge weights c that define the capacity of each edge. We can identify two special nodes in G: the source node s and the sink node t. We want to find

More information

Parallel static and dynamic multi-constraint graph partitioning

Parallel static and dynamic multi-constraint graph partitioning CONCURRENCY AND COMPUTATION: PRACTICE AND EXPERIENCE Concurrency Computat.: Pract. Exper. 2002; 14:219 240 (DOI: 10.1002/cpe.605) Parallel static and dynamic multi-constraint graph partitioning Kirk Schloegel,,

More information

Hypergraph-Partitioning-Based Decomposition for Parallel Sparse-Matrix Vector Multiplication

Hypergraph-Partitioning-Based Decomposition for Parallel Sparse-Matrix Vector Multiplication IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, VOL. 10, NO. 7, JULY 1999 673 Hypergraph-Partitioning-Based Decomposition for Parallel Sparse-Matrix Vector Multiplication UÈ mitv.cëatalyuè rek and

More information

Scalable Clustering of Signed Networks Using Balance Normalized Cut

Scalable Clustering of Signed Networks Using Balance Normalized Cut Scalable Clustering of Signed Networks Using Balance Normalized Cut Kai-Yang Chiang,, Inderjit S. Dhillon The 21st ACM International Conference on Information and Knowledge Management (CIKM 2012) Oct.

More information

Using Multi-Level Graphs for Timetable Information. Frank Schulz, Dorothea Wagner, and Christos Zaroliagis

Using Multi-Level Graphs for Timetable Information. Frank Schulz, Dorothea Wagner, and Christos Zaroliagis Using Multi-Level Graphs for Timetable Information Frank Schulz, Dorothea Wagner, and Christos Zaroliagis Overview 1 1. Introduction Timetable Information - A Shortest Path Problem 2. Multi-Level Graphs

More information

Multi-Resource Aware Partitioning Algorithms for FPGAs with Heterogeneous Resources

Multi-Resource Aware Partitioning Algorithms for FPGAs with Heterogeneous Resources Multi-Resource Aware Partitioning Algorithms for FPGAs with Heterogeneous Resources Navaratnasothie Selvakkumaran Abhishek Ranjan HierDesign Inc Salil Raje HierDesign Inc George Karypis Department of Computer

More information

EE 701 ROBOT VISION. Segmentation

EE 701 ROBOT VISION. Segmentation EE 701 ROBOT VISION Regions and Image Segmentation Histogram-based Segmentation Automatic Thresholding K-means Clustering Spatial Coherence Merging and Splitting Graph Theoretic Segmentation Region Growing

More information

Graph Coarsening and Clustering on the GPU

Graph Coarsening and Clustering on the GPU Graph Coarsening and Clustering on the GPU B. O. Fagginger Auer R. H. Bisseling Utrecht University February 14, 2012 Introduction We will discuss coarsening and greedy clustering of graphs. Introduction

More information

PT-Scotch: A tool for efficient parallel graph ordering

PT-Scotch: A tool for efficient parallel graph ordering PT-Scotch: A tool for efficient parallel graph ordering C. Chevalier a, F. Pellegrini b a LaBRI & Project ScAlApplix of INRIA Futurs 351, cours de la Libération, 33400 Talence, France b ENSEIRB, LaBRI

More information

Evolving Multi-level Graph Partitioning Algorithms

Evolving Multi-level Graph Partitioning Algorithms Evolving Multi-level Graph Partitioning Algorithms Aaron S. Pope, Daniel R. Tauritz and Alexander D. Kent Department of Computer Science, Missouri University of Science and Technology, Rolla, Missouri

More information

Better Process Mapping and Sparse Quadratic Assignment

Better Process Mapping and Sparse Quadratic Assignment Better Process Mapping and Sparse Quadratic Assignment Christian Schulz 1 and Jesper Larsson Träff 2 1 Karlsruhe Institute of Technology, Karlsruhe, Germany; and University of Vienna, Vienna, Austria christian.schulz@kit.edu,

More information

Scalable, Hybrid-Parallel Multiscale Methods using DUNE

Scalable, Hybrid-Parallel Multiscale Methods using DUNE MÜNSTER Scalable Hybrid-Parallel Multiscale Methods using DUNE R. Milk S. Kaulmann M. Ohlberger December 1st 2014 Outline MÜNSTER Scalable Hybrid-Parallel Multiscale Methods using DUNE 2 /28 Abstraction

More information

Multilevel k-way Hypergraph Partitioning

Multilevel k-way Hypergraph Partitioning _ Multilevel k-way Hypergraph Partitioning George Karypis and Vipin Kumar fkarypis, kumarg@cs.umn.edu Department of Computer Science & Engineering, University of Minnesota, Minneapolis, MN 55455 Abstract

More information

Preclass Warmup. ESE535: Electronic Design Automation. Motivation (1) Today. Bisection Width. Motivation (2)

Preclass Warmup. ESE535: Electronic Design Automation. Motivation (1) Today. Bisection Width. Motivation (2) ESE535: Electronic Design Automation Preclass Warmup What cut size were you able to achieve? Day 4: January 28, 25 Partitioning (Intro, KLFM) 2 Partitioning why important Today Can be used as tool at many

More information

Exploring unstructured Poisson solvers for FDS

Exploring unstructured Poisson solvers for FDS Exploring unstructured Poisson solvers for FDS Dr. Susanne Kilian hhpberlin - Ingenieure für Brandschutz 10245 Berlin - Germany Agenda 1 Discretization of Poisson- Löser 2 Solvers for 3 Numerical Tests

More information

Multiresolution Computation of Conformal Structures of Surfaces

Multiresolution Computation of Conformal Structures of Surfaces Multiresolution Computation of Conformal Structures of Surfaces Xianfeng Gu Yalin Wang Shing-Tung Yau Division of Engineering and Applied Science, Harvard University, Cambridge, MA 0138 Mathematics Department,

More information

CS 534: Computer Vision Segmentation II Graph Cuts and Image Segmentation

CS 534: Computer Vision Segmentation II Graph Cuts and Image Segmentation CS 534: Computer Vision Segmentation II Graph Cuts and Image Segmentation Spring 2005 Ahmed Elgammal Dept of Computer Science CS 534 Segmentation II - 1 Outlines What is Graph cuts Graph-based clustering

More information

Compressing Social Networks

Compressing Social Networks Compressing Social Networks The Minimum Logarithmic Arrangement Problem Chad Waters School of Computing Clemson University cgwater@clemson.edu March 4, 2013 Motivation Determine the extent to which social

More information

A GPU Algorithm for Greedy Graph Matching

A GPU Algorithm for Greedy Graph Matching A GPU Algorithm for Greedy Graph Matching B. O. Fagginger Auer R. H. Bisseling Utrecht University September 29, 2011 Fagginger Auer, Bisseling (UU) GPU Greedy Graph Matching September 29, 2011 1 / 27 Outline

More information