Improving parallel execution performance for logic programs using mode information

Parallel execution of a logic program is an active research topic. Extensive study has been focused on the AND-OR parallel execution of a logic program. Current approaches adopt a top-down evaluation sequence to realize AND-OR parallelism by exploring parallelly executable predicates from root to leaf and then collect bindings generated in this process. Our research presented in this paper explores the parallel execution of a logic program based on a new data flow analysis algorithm. A logic program is first applied by data dependency analysis which can find all mode combinations possibly existing within a logic clause. This mode information is used to support a novel hybrid parallel execution model, which combines both top-down and bottom-up evaluation strategies. By adopting this model, various improvements can be achieved. The results generated from a simulator demonstrate very encouraging results.<
>