Aim.The introduction of the full paper discusses what we believe to be the shortcomings of a number of papers in the open literature about real-time scheduling algorithms and then, in its last paragraph, proposes what we believe to be a new and better DTSA, which is explained in sections 1 and 2.Section 1 gives a dependable scheduling model which considers both reliability and security and is suitable for the distributed real-time system that has high requirements for dependability.Section 2 proposes our improved DTSA which combines fault-tolerant technology with information security technology; it uses the improved backup copy technology to effectively assign and schedule real-time tasks among many processors, thus enhancing the utilization of CPU resources.It also supports to handle hardware failure and software failure.Specially, it prevents these malice tasks from running on processor by checking its identify before running.Section 3 simulates our DTSA; the simulation results, presented in Figs.1, and their analysis show preliminarily that, compared with other algorithms, our DSTA can obtain higher performance with little cost in hardware, thus having better dependability.%目前许多面向关键领域的应用对实时系统提出了可信要求,研究可信的实时调度算法成为当前的研究热点.文章分析了当前公开文献中各种实时调度算法的缺陷,针对现有算法的不足,提出了一种适用于分布式实时系统的启发式可信调度算法,该算法以提高系统的可靠性和安全性为目标,一方面采用改进的主动副本复制技术,在多个处理机上有效分配实时周期任务,并支持对硬件失效和软件瞬时失效的处理,另一方面利用处理机对每个运行之前的任务进行安全性检查,有效防止了非法任务进入并破坏系统.仿真实验的结果表明,文中提出的算法DTSA相对于其它算法而言,在仅增加较少硬件成本的前提下,能够有效提高运行任务的可靠性和安全性.
展开▼
