欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
传统的启发式算法解决容错实时任务调度时,在性能和系统利用率方面表现得不太理想,提出一种新的基于主副版本实时任务容错机制的帝国竞争算法.该算法把容错知识和帝国竞争算法相融合,利用适应度共享技术对国家的适应度进行调整,以提高全局寻优能力,有效避免算法早熟收敛.实验结果表明：与同类算法相比,该算法在系统利用率和效率上表现出较好的特性.
英文摘要：
Conventional methods for fault-tolerant scheduling of real-time tasks based on heuristic approach offer poor performance and inefficient system utilization. The primary-backup （PB） approach is often used as a fault tolerant scheduling technique to guarantee real-time （RT） tasks to meet their deadline. We propose a novel scheduling algorithm using imporoved discrete imperialistic competition algorithm （IDICA） hybridized with knowledge from the real-time task scheduling domain for providing fault-tolerance （FT） in multiprocessor environment. Exhaustive simulation reveals that the IDICA based primary backup fault tolerant scheduling approach outperforms other fault tolerant scheduling schemes in terms of system utilization and efficiency.
同期刊论文项目
同项目期刊论文
期刊信息
位置：