计算机应用 ›› 2005, Vol. 25 ›› Issue (04): 960-961.DOI: 10.3724/SP.J.1087.2005.0960

• 典型应用 • 上一篇    下一篇

基于非精确计算的多错误模式下的容错优化调度算法

罗琼,张立臣   

  1. 广东工业大学计算机学院
  • 发布日期:2005-04-01 出版日期:2005-04-01
  • 基金资助:

    国家自然科学基金资助项目(60174050;60474072);;广东省自然科学基金 资助项目(4009465;010059);;广东省高校自然科学研究项目(Z03024)

Imprecise-based fault-tolerant optimal scheduling algorithm in multi-error model

LUO Qiong,ZHANG Li-chen   

  1. Faculty of Computer,Guangdong University of Technology
  • Online:2005-04-01 Published:2005-04-01

摘要: 实时系统中每个任务必须在其截止时间内产生逻辑正确结果。但是,由于过载并不保证 每个实时任务都能满足此要求。因此,适当降低实时任务精确度以及在错误发生时提供最大利用率 的容错优化方法以保证系统的实时性与可靠性变得非常重要。利用非精确调度中任务的可选部分为 强制部分的错误恢复提供潜在的时间容余,提出了在多个错误发生时既能保证报酬最大化又能提供 强制部分恢复处理的容错优化调度形式算法,并结合实例给出分析过程。

关键词: 非精确计算, 容错优化调度, k错误模式的容错优化算法

Abstract:  In a real-time system, each task must complete and produce correct output by the specified deadline. But, it is not possible to meet each deadline because of system overloaded. So,it is very important to decrease precision of real-time task and provide a maximize-utility fault-tolerant optimal scheduling to assure timeliness and reliability in the presence of multi-error. The optional parts of task in imprecise computation was used to provide potential time redundant for error-recovery of mandatory parts, and a formal k-fault tolerant optimal scheduling algorithm was proposed, which can guarantee maximize-reward and provide recovery operation for mandatory simultaneous. The analysis process was given by an example.

Key words:  imprecise computation, fault-tolerant optimal scheduling, k-fault tolerant optimal algorithm

中图分类号: