Towards a complete constraint solver on GPU

ABSTRACT. Constraint programming has gained prominence as an effective and declarative paradigm for modeling and solving complex combinatorial problems. In spite of the natural presence of concurrency, there has been relatively limited effort to use novel massively parallel architectures-such as those found in modern Graphical Processing Units (GPUs)-to speedup constraint programming algorithms. This paper presents an ongoing work for the development of a constraint solver which exploits GPU parallelism. The work is based on two previous results where constraint propagation and approximated search have been parallelized. We summarize these result and discuss the features we have planned to carry on.