Parallel Execution Models for Constraint Propagation

2011

With the increased availability of affordable parallel and dis-tributed hardware, programming models for these architectures has be-come the focus of significant attention. Constraint programming, which can be seen as the encoding of processes as a Constraint Satisfaction Problem, because of its data-driven and control-insensitive approach is a prime candidate to serve as the basis for a framework which effectively exploits parallel architectures.

Lecture Notes in Computer Science, 1999

Many problems from artificial intelligence can be described as constraint satisfaction problems over finite domains (CSP(FD)), that is, a solution is an assignment of a value to each problem variable such that a set of constraints is satisfied. Arc-consistency algorithms remove inconsistent values from the set of values that can be assigned to a variable (its domain), thus reducing the search space. We have developed a parallelisation scheme of arc-consistency to be run on MIMD multiprocessor. The set of constraints is divided into N partitions, which are executed in parallel on N processors. The parallelisation scheme has been implemented on a CRAY T3E multiprocessor with up to thirty-four processors. Empirical results on speedup and behaviour are reported and discussed.

PDPTA

Abstract— Program parallelization becomes increasingly important when new multi-core architectures provide ways to improve performance. One of the greatest challenges of this development lies in programming parallel applications. Us-ing declarative languages, such as constraint ...

This article presents the qualitative and quantitative results of experiments applying parallelism to constraint programming. We do not set out to prove that parallel constraint programming is the perfect problem-solving tool, but rather to characterize what can be expected from this combination and to discover when using parallelism is productive. In these experiments, we do not deal with scalability involving large numbers of processors; nor do we try to solve extremely difficult or large problems by using brute force algorithms running for days, weeks or even longer. Instead, we concentrate on parallelism involving one to four processors and we restrict our solving time to small durations (around 15 minutes in general). This protocol is applied to different series of problems. The results are analyzed in order to gain a better understanding of the practical benefits of parallel constraint programming.

1996

This paper describes the rst results from research on the compilation of constraint systems into task level procedural parallel programs. Algorithms are expressed as constraint systems. A data ow graph is derived from the constraint system and a set of input variables. The data ow graph, which exploits the parallelism in the constraints, is mapped to the target language CODE 2.0, which represents parallel computation structures as generalized dependence graphs. Finally, parallel C programs are generated. The granularity of the derived data ow graphs depends upon the complexity of the operations directly represented in the constraint system. To extract parallel programs of appropriate granularity, the following features have been included: (i) modularity, (ii) operations over structured types as primitives, (iii) de nition of sequential C functions. A prototype of the compiler has been implemented. The domain of matrix computations has been targeted for applications. Some examples have been programmed. Initial results are encouraging.

International Conference on Logic Programming/Joint International Conference and Symposium on Logic Programming, 1998

We present in this paper a parallel execution model of arc-consistency for Constraint SatisfactionProblems (CSP), implemented on a scaleable MIMD distributed memory machine.We have adopted the indexical scheme, an adequate approach to arc-consistency for functionalconstraints. The CSP is partitioned into N partitions, which are executed in parallel on N processors.Each processor applies sequential arc-consistency to its subset of constraints, updatingremote

2007

Abstract. Constraint programming libraries are useful when building applications developed mostly in mainstream programming languages: they do not require the developers to acquire skills for a new language, providing instead declarative programming tools for use within conventional systems. Some approaches to constraint programming favour completeness, such as propagation-based systems.

Parallel Algorithms and Applications, 1995

This paper deals with parallel checking of passive constraints in object-oriented databases. It presents a parallel algorithm for constraint checking based on a master-slave technique and discusses its implementation on a parallel object-oriented database system. The system is named PRACTIC and is based on class concurrency. Passive constraints, unlike active database rules, are independent and can be executed using AND-parallelism. Simulation shows that the proposed algorithm offers considerable speed-up, which mainly depends on the number of constraints and the total constraint execution time, while it is only slightly affected from the distribution of constraints and the constraint scheduling policy. Finally, it is explained how the PRACTIC system enhances the algorithm's performance using features, like nested query parallelism and constraint overlapping.

Lecture Notes in Computer Science, 2004

As a case study that illustrates our view on coordination and component-based software engineering, we present the design and implementation of a parallel constraint solver. The parallel solver coordinates autonomous instances of a sequential constraint solver, which is used as a software component. The component solvers achieve load balancing of tree search through a time-out mechanism. Experiments show that the purely exogenous mode of coordination employed here yields a viable parallel solver that effectively reduces turn-around time for constraint solving on a broad range of hardware platforms.

Theory and Practice of Logic Programming, 2011

We introduce a systematic, concurrent execution scheme for Constraint Handling Rules (CHR) based on a previously proposed sequential goal-based CHR semantics. We establish strong correspondence results to the abstract CHR semantics, thus guaranteeing that any answer in the concurrent, goal-based CHR semantics is reproducible in the abstract CHR semantics. Our work provides the foundation to obtain efficient, parallel CHR execution schemes.