Planning of parallel abstract programs as boolean satisfiability

Authors:
Gennady A. Oparin;Alexei P. Novopashin
Affiliations:
Institute of Systems Dynamics and Control Theory, Siberian Branch of Russian Academy of Sciences, Irkutsk, Russia;Institute of Systems Dynamics and Control Theory, Siberian Branch of Russian Academy of Sciences, Irkutsk, Russia
Venue:
PaCT'05 Proceedings of the 8th international conference on Parallel Computing Technologies
Year:
2005

Citing 1
Cited 0

Planning as satisfiability

ECAI '92 Proceedings of the 10th European conference on Artificial intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, a new formulation is proposed for the problem of constructing parallel abstract programs of a required length in parallel computing systems. The conditions of a planning problem are represented as a system of Boolean equations (constraints), whose solutions determine the possible plans for activating the program modules. Specifications of modules are stored in the knowledgebase of the planner. Constraint on number of processors and time delays at execution of modules are taken into account.