Parallel Logic Programming for Problem Solving

  • Authors:
  • Ramiro Varela Arias;Camino Rodríguez Vela;Jorge Puente Peinador;Cesar Alonso González

  • Affiliations:
  • Centro de Inteligencia Artificial, Universidad de Oviedo en Gijón Campus de Viesques, E-33271 Gijón, Spain. {ramiro,camino,puente,calonso}@aic.uniovi.esht ...;Centro de Inteligencia Artificial, Universidad de Oviedo en Gijón Campus de Viesques, E-33271 Gijón, Spain. {ramiro,camino,puente,calonso}@aic.uniovi.esht ...;Centro de Inteligencia Artificial, Universidad de Oviedo en Gijón Campus de Viesques, E-33271 Gijón, Spain. {ramiro,camino,puente,calonso}@aic.uniovi.esht ...;Centro de Inteligencia Artificial, Universidad de Oviedo en Gijón Campus de Viesques, E-33271 Gijón, Spain. {ramiro,camino,puente,calonso}@aic.uniovi.esht ...

  • Venue:
  • International Journal of Parallel Programming
  • Year:
  • 2000

Quantified Score

Hi-index 0.00

Visualization

Abstract

We present a new model for parallel evaluation of logic programs. This model can exploit the main sources of parallelism that the language of logic expresses: Independent {\sc and} parallelism and {\sc or} parallelism, together with a secondary source emerging as a consequence of the Independent {\sc and} Parallelism: the producer/consumer parallelism. The efficiency is derived from the use of ordered structures for managing the information generated throughout the search process. The model is suitable for evaluating programs with a high degree of non-determinism because it never generates two processes for solving the same subgoal and hence it can exploit the same real parallelism generating a lower number of processes than other models. As an application example, we consider the Job Shop Scheduling problem. We report experimental results showing that logic programs can be designed that exhibit parallelism, and that the use of heuristic information translates into speedup in obtaining answers.