Partitioning Methods to Parallelize Constraint Programming Solver Using the Parallel Framework Bobpp

Menouer, Tarek; Cun, Bertrand Le; Vander-Swalmen, Pascal

doi:10.1007/978-3-319-00293-4_10

Cited by 8 publications

(7 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The pending threads take effect by the insertion of a new OR-Tools-Bob-node in the priority queue. This method of partitioning is extensively studied in [16].…”

Section: A Porting Of Or-tools Solver On Top Of Bobppmentioning

confidence: 99%

“…As OR-Tools solver run in sequential, we present also the principle of method used to parallelize the ORTools search space using the parallel Bobpp framework [12], [16], this method is based on Work Stealing technique in order to partition the search-tree in a set of sub-trees and assigns each sub-tree to one computing core during the execution of the algorithm.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Parallelization Mixing OR-Tools/Gecode Solvers on Top of the Bobpp Framework

Menouer

Cun

2013

2013 Eighth International Conference on P2P, Parallel, Grid, Cloud and Internet Computing

Self Cite

View full text Add to dashboard Cite

The novelty of this paper is to propose a parallel solver that mixes two Constraint Programming (CP) solvers, on top of our Bobpp framework, in order to have a parallel CP solver portfolio. The first chosen CP solver is OR-tools which is an open source sequential solver. We propose an external ORTools parallelization (partitioning/migration/load balancing) on top of the Bobpp framework, where different instances of the OR-Tools solver are executed on different cores. The second chosen CP solver is Gecode that is also an open source sequential and parallel solver. The Gecode parallelization is internal in a sense that the parallel strategy is included in the library. The advantage of Gecode is that it can be easily interfaced with other software. Bobpp is a parallel framework that provides an interface between: Combinatorial Optimization problems or CP solvers and the parallel machines. Bobpp is designed to propose several parallelizations according to the parallel programming environment and the architecture of a machine. The performances of our solver portfolio are illustrated by solving CP problems modeled in FlatZinc format.

show abstract

“…The pending threads take effect by the insertion of a new OR-Tools-Bob-node in the priority queue. This method of partitioning is extensively studied in [16].…”

Section: A Porting Of Or-tools Solver On Top Of Bobppmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A Parallelization Mixing OR-Tools/Gecode Solvers on Top of the Bobpp Framework

Menouer

Cun

2013

2013 Eighth International Conference on P2P, Parallel, Grid, Cloud and Internet Computing

Self Cite

View full text Add to dashboard Cite

show abstract

“…Another drawback is that the implementation of the work stealing is intrusive and is strongly dependent of the solver which requires to have a very good knowledge of the CP solver and to access to some internal functions. Some methods try to address this issue [7].…”

Section: Work Stealingmentioning

confidence: 99%

Improvement of the Embarrassingly Parallel Search for Data Centers

Régin

Rezgui

Malapert

2014

Lecture Notes in Computer Science

View full text Add to dashboard Cite

International audienceWe propose an adaptation of the Embarrassingly Parallel Search (EPS) method for data centers. EPS is a simple but efficient method for parallel solving of CSPs. EPS decomposes the problem in many distinct subproblems which are then solved independently by workers. EPS performed well on multi-cores machines (40), but some issues arise when using more cores in a datacenter. Here, we identify the decomposition as the cause of the degradation and propose a parallel decomposition to address this issue. Thanks to it, EPS gives almost linear speedup and outperforms work stealing by orders of magnitude using the Gecode solver

show abstract

“…The tree partitioning may be static or dynamic using Work Stealing technique. In the most studies performed to parallelize search spaces as studies used in CP [11], [14], [7], [16], [6], the load balancing procedure has full control of the search. • A portfolio parallelization, the principle is to run N search strategies using N computing cores.…”

Section: Introductionmentioning

confidence: 99%

Adaptive N to P Portfolio for Solving Constraint Programming Problems on Top of the Parallel Bobpp Framework

Menouer

Cun

2014

2014 IEEE International Parallel &Amp; Distributed Processing Symposium Workshops

Self Cite

View full text Add to dashboard Cite

This paper presents a parallelization of Constraint Programming (CP) solver, based on the portfolio principle, in order to quickly solve constraint satisfaction and optimisation problems. The portfolio principle is widely used in the parallelization of boolean SATisfiability (SAT) and CP solvers. It is based on the running of N search strategies for the same problem using N computing cores. Each core uses its own strategy in order to perform a search that is different form the other ones. The first strategy that responds to the needs of the user stops all other strategies.In the usual portfolio principle, the number of search strategies is limited compared to the current number of the computing cores used by parallel machines. The idea of this article is to run N search strategies for the same CP problem and schedule these strategies using P computing cores (P > N ). The novelty is that the scheduling of these N strategies is dynamically performed between the different computing cores. The goal is to adapt the scheduling of the search strategies so as to favour the strategy that finds a solution quickly. The performances obtained with this adaptive portfolio solver are illustrated by solving the CP problems modeled using FlatZinc format and solved using the OR-Tools solver on top of the parallel Bobpp framework.

show abstract

Partitioning Methods to Parallelize Constraint Programming Solver Using the Parallel Framework Bobpp

Cited by 8 publications

References 13 publications

A Parallelization Mixing OR-Tools/Gecode Solvers on Top of the Bobpp Framework

A Parallelization Mixing OR-Tools/Gecode Solvers on Top of the Bobpp Framework

Improvement of the Embarrassingly Parallel Search for Data Centers

Adaptive N to P Portfolio for Solving Constraint Programming Problems on Top of the Parallel Bobpp Framework

Contact Info

Product

Resources

About