Scalable Parallel Implementation of Constraint Programs on Distributed Network of SMPs

Efficient scalable implementations of Constraint Logic Programming Systems are being designed. To date, a model based on stack-splitting has been proposed to realize efficient distributed implementations of constraint-based system. The implementation of this model on a distributed network of SMPs on top of a sequential engine from ALS, Inc., is currently in progress. Other NMSU people involved are Prof. Enrico Pontelli, Hai-Feng Guo, and Karen Villaverde. The project is being done in collaboration with ALS, Inc., Federal University of Rio de Janeiro and Federal University of Rio Grande do Sul (Brazil), University of Porto (Portugal) and is funded by NSF and a Department of Education graduate fellowship.

Papers

G. Gupta and E. Pontelli. Stack-splitting: A Simple Technique for Implementing Or-parallelism and And-parallelism on Distributed Machines. In Proc. 16th International Conference on Logic Programming, 1999. MIT Press, pp. 290-305.
G. Gupta, E. Pontelli. Efficient Techniques for Distributed Implementation of Search-based AI Systems. In International Conference on Parallel Processing, 1999 (Aizu, Japan). pp. 319-326.
H-F. Guo, K. Villaverde, E. Pontelli, G. Gupta. Cuts and Side-Effects in a Distributed Or-parallel Implementation. Working paper.
E. Pontelli, G. Gupta. A Simulation Study of Distributed Execution of (Constraint) Logic Programs. Working Paper.