Parallel Discrepancy-Based Search

Moisan, Thierry; Gaudreault, Jonathan; Quimper, Claude-Guy

doi:10.1007/978-3-642-40627-0_6

Cited by 18 publications

(11 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Secondly, they claim that "for many problems the heuristics are least reliable early in the search, before making decisions that reduce the problem to a size for which the heuristics become reliable". There is a long-standing tradition of focusing upon the first of these two claims, and ignoring the second (Korf, 1996;Walsh, 1997;Prosser and Unsworth, 2011;Moisan et al, 2013). Here we will buck the trend and emphasise the second claim.…”

Section: The Quality Of Heuristics and What This Impliesmentioning

confidence: 94%

The Shape of the Search Tree for the Maximum Clique Problem and the Implications for Parallel Branch and Bound

McCreesh

Prosser

2015

ACM Trans. Parallel Comput.

View full text Add to dashboard Cite

Finding a maximum clique in a given graph is one of the fundamental NP-hard problems. We compare two multi-core thread-parallel adaptations of a state-of-the-art branch and bound algorithm for the maximum clique problem, and provide a novel explanation as to why they are successful. We show that load balance is sometimes a problem, but that the interaction of parallel search order and the most likely location of solutions within the search space is often the dominating consideration. We use this explanation to propose a new low-overhead, scalable work splitting mechanism. Our approach uses explicit early diversity to avoid strong commitment to the weakest heuristic advice, and late resplitting for balance. More generally, we argue that for branch and bound, parallel algorithm design should not be performed independently of the underlying sequential algorithm.

show abstract

Section: The Quality Of Heuristics and What This Impliesmentioning

confidence: 94%

The Shape of the Search Tree for the Maximum Clique Problem and the Implications for Parallel Branch and Bound

McCreesh

Prosser

2015

ACM Trans. Parallel Comput.

View full text Add to dashboard Cite

show abstract

“…A similar strategy is exploited in [11], where however the details of the split are very much dependent of the LDS algorithm, and cannot be easily (and efficiently) generalized to an arbitrary tree search algorithm.…”

Section: Vanilla Algorithmmentioning

confidence: 99%

“…All subproblems are put into a queue and distributed to workers as needed (usually, a subproblem is assigned to a given worker as soon as the worker is idle). In [11], a parallelization strategy for LDS [12] is presented, in which the leaves of the complete LDS tree are deterministically assigned to the workers, and each worker processes a subtree only if it contains a leaf assigned to it. These strategies share some similarities with our approach, although some important differences remain.…”

Section: Introductionmentioning

confidence: 99%

Self-splitting of Workload in Parallel Computation

Fischetti¹,

Monaci²,

Salvagnin³

2014

Integration of AI and OR Techniques in Constraint Programming

View full text Add to dashboard Cite

Abstract. Parallel computation requires splitting a job among a set of processing units called workers. The computation is generally performed by a set of one or more master workers that split the workload into chunks and distribute them to a set of slave workers. In this setting, communication among workers can be problematic and/or time consuming. Tree search algorithms are particularly suited for being applied in a parallel fashion, as different nodes can be processed by different workers in parallel. In this paper we propose a simple mechanism to convert a sequential tree-search code into a parallel one. In the new paradigm, called SelfSplit, each worker is able to autonomously determine, without any communication with the other workers, the job parts it has to process. Computational results are reported, showing that SelfSplit can achieve an almost linear speedup for hard Constraint Programming applications, even when 64 workers are considered.

show abstract

“…A simple constraint solving method for project scheduling problems has also been implemented on an IBM Bluegene/P supercomputer [77] up to 1,024 cores, but with mixed results since they reach linear speedups until 512 cores only, with no improvements beyond this limit. Very recently another optimization method, Limited Discrepancy Search, has been parallelized with very good results up to a few thousands of cores [57]. An important point is that the proposed method does not requires communication between parallel processes and has good load-balancing properties.…”

Section: Introductionmentioning

confidence: 99%

Large-scale parallelism for constraint-based local search: the costas array case study

et al. 2014

View full text Add to dashboard Cite

We present the parallel implementation of a constraint-based Local Search algorithm and investigate its performance on several hardware platforms with several hundreds or thousands of cores. We chose as the basis for these experiments the Adaptive Search method, an efficient sequential Local Search method for Constraint Satisfaction Problems (CSP). After preliminary experiments on some CSPLib benchmarks, we detail the modeling and solving of a hard combinatorial problem related to radar and sonar applications: the Costas Array Problem. Performance evaluation on some classical CSP benchmarks shows that speedups are very good for a few tens of cores, and good up to a few hundreds of cores. However for a hard combinatorial search problem such as the Costas Array Problem, performance evaluation of the sequential version shows results outperforming previous Local Search implementations, while the parallel version shows nearly linear speedups up to 8,192 cores. The proposed parallel scheme is simple and based on independent multi-walks with no communication between processes during search. We also investigated a cooperative multi-walk scheme where processes share simple information, but this scheme does not seem to improve performance.

show abstract

Parallel Discrepancy-Based Search

Cited by 18 publications

References 19 publications

The Shape of the Search Tree for the Maximum Clique Problem and the Implications for Parallel Branch and Bound

The Shape of the Search Tree for the Maximum Clique Problem and the Implications for Parallel Branch and Bound

Self-splitting of Workload in Parallel Computation

Large-scale parallelism for constraint-based local search: the costas array case study

Contact Info

Product

Resources

About