A GPU Implementation of Parallel Constraint-Based Local Search

In this paper we study the performance of constraint-based local search solvers on a GPU. The massively parallel architecture of the GPU makes it possible to explore parallelism at two different levels inside the local search algorithm. First, by executing multiple copies of the algorithm in a multi-walk manner and, second, by evaluating large neighborhoods in parallel in a single-walk manner. Experiments on three well-known problem benchmarks indicate that the current GPU implementation is up to 17 times faster than a well-tuned sequential algorithm implemented on a desktop computer.

[1]  Jin-Kao Hao,et al.  A General Approach for Constraint Solving by Local Search , 2004, J. Math. Model. Algorithms.

[2]  Christian Schulz,et al.  Efficient local search on the GPU - Investigations on the vehicle routing problem , 2013, J. Parallel Distributed Comput..

[3]  Philippe Codognet,et al.  Using sequential runtime distributions for the parallel speedup prediction of SAT local search , 2013, Theory and Practice of Logic Programming.

[4]  Philippe Codognet,et al.  An Efficient Library for Solving CSP with Local Search , 2003 .

[5]  Philippe Codognet,et al.  Parallel Local Search for the Costas Array Problem , 2012, 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops & PhD Forum.

[6]  Matheus Nohra Haddad,et al.  A Hybrid CPU-GPU Local Search Heuristic for the Unrelated Parallel Machine Scheduling Problem , 2012, 2012 Third Workshop on Applications for Multi-Core Architecture.

[7]  El-Ghazali Talbi,et al.  GPU Computing for Parallel Local Search Metaheuristic Algorithms , 2013, IEEE Transactions on Computers.

[8]  Philippe Codognet,et al.  Experiments in Parallel Constraint-Based Local Search , 2011, EvoCOP.

[9]  Alessandro Dal Palù,et al.  Exploiting Unexploited Computing Resources for Computational Logics , 2012, CILC.

[10]  Michael J. Maher,et al.  Kangaroo: An Efficient Constraint-Based Local Search System Using Lazy Propagation , 2011, CP.

[11]  Philippe Codognet,et al.  Yet Another Local Search Method for Constraint Solving , 2001, SAGA.

[12]  Jakub Marecek,et al.  Handbook of Approximation Algorithms and Metaheuristics , 2010, Comput. J..

[13]  Philippe Codognet,et al.  Parallel Constraint-Based Local Search on the Cell/BE Multicore Architecture , 2010, IDC.

[14]  Philippe Codognet,et al.  Constraint-Based Local Search for the Costas Array Problem , 2012, LION.

[15]  Daniel Diaz,et al.  Parallel Local Search: Experiments with a PGAS-based programming model , 2013, ArXiv.

[16]  Martin Burtscher,et al.  A Scalable Heterogeneous Parallelization Framework for Iterative Local Searches , 2013, 2013 IEEE 27th International Symposium on Parallel and Distributed Processing.

[17]  Panagiotis Manolios,et al.  Implementing Survey Propagation on Graphics Processing Units , 2006, SAT.

[18]  Philippe Codognet,et al.  Towards Massively Parallel Local Search for SAT - (Poster Presentation) , 2012, SAT.

[19]  Emile H. L. Aarts,et al.  Parallel local search , 1995, J. Heuristics.

[20]  Guy E. Blelloch,et al.  Scans as Primitive Parallel Operations , 1989, ICPP.

[21]  El-Ghazali Talbi,et al.  Metaheuristics on GPUs , 2013, J. Parallel Distributed Comput..

[22]  K. Drakakis,et al.  A review of Costas arrays , 2006 .

[23]  Steven Minton,et al.  Minimizing Conflicts: A Heuristic Repair Method for Constraint Satisfaction and Scheduling Problems , 1992, Artif. Intell..

[24]  Philippe Codognet,et al.  Prediction of Parallel Speed-Ups for Las Vegas Algorithms , 2012, 2013 42nd International Conference on Parallel Processing.

[25]  C.J.H. Mann,et al.  Handbook of Approximation: Algorithms and Metaheuristics , 2008 .

[26]  Serdar Kadioglu,et al.  Dialectic Search , 2009, CP.

[27]  Pierre Hansen,et al.  Cooperative Parallel Variable Neighborhood Search for the p-Median , 2004, J. Heuristics.