讨论一组实时系统的任务在不同性质的处理器上的分配,使得所有任务得以完成并耗费更少的时间,是NP完全问题。建立了新的任务分析模型——异构多核多帧任务模型,并基于遗传算法给出解决方案。实验证明,该模型更为有效地表达了实时系统的性质,获得更高的分配成功率,算法拥有更低的时间复杂度,结果可信。
Given a set of tasks and a collection of different kind processors, the problem Was determining whether the tasks could be partitioned among the processors in such a manner that all timing constraints were met. This problem was intractable. A new task model: built heterogeneous multi-proeessors multi-frame task model and solution in genetie algorithms. Result shows that new model is more generalized, achieves larger succeed partitioning percent; the algorithms have lower time-complexity and reliable result.