作者: Steve Kuo , Dan Moldovan
DOI: 10.1016/0743-7315(91)90098-T
关键词: Intel iPSC 、 Hypercube 、 Parallel computing 、 Computer science
摘要: Abstract The performance of production programs can be improved by firing multiple rules concurrently at a cycle. Although considerable amount research has been done on parallel processing programs, the problem rule not thoroughly investigated yet. In this paper, we begin identifying problems associated with systems, compatibility and convergence problems, present three models which address them. dependence model addresses using interrule data analysis. single-context-multiple-rules (SCMR) multiple-contexts-multiple-rules (MCMR) both problems. A program executed under SCMR MCMR reaches solution is equivalent to sequential execution. These have simulated RUBIC simulator, model, highest performance, implemented Intel iPSC/2 hypercube. simulation implementation results are reported.