将现有入侵容忍、自毁技术与自律计算相结合,提出了一种基于SM-PEPA(semi-Markov performance evaluation process algebra)的关键任务系统自律可信性模型以支持形式化分析和推理.该模型具有一定程度的自管理能力,采用分级处理的方式应...将现有入侵容忍、自毁技术与自律计算相结合,提出了一种基于SM-PEPA(semi-Markov performance evaluation process algebra)的关键任务系统自律可信性模型以支持形式化分析和推理.该模型具有一定程度的自管理能力,采用分级处理的方式应对各种程度的可信性威胁,满足了关键任务系统对可信性的特殊需求.在此基础上,从稳态概率角度提出了一种自律可信性度量方法.最后,结合具体实例对模型参数对自律可信性的影响进行了初步分析.实验结果表明,增大关键任务系统可信性威胁检测率和自恢复成功率,可在较大范围内提高系统的自律可信特性.展开更多
The software architecture is a design of application system,performance evaluation of software architecture during the early stage of their development is really attractive. This paper proposes a new method for softwa...The software architecture is a design of application system,performance evaluation of software architecture during the early stage of their development is really attractive. This paper proposes a new method for software architecture performance modeling. To achieve this ,it adds new calculus into stochastic process algebra(SPA in short),the developed SPA called extended stochastic process algebra (ESPA in short). By ESPA,performance evaluation and software architecture modeling can combine perfectly. It defines a few performance terms for software architecture using reward structure derived from ESPA. To explain the performance terms it also designs an experiment.展开更多
文摘将现有入侵容忍、自毁技术与自律计算相结合,提出了一种基于SM-PEPA(semi-Markov performance evaluation process algebra)的关键任务系统自律可信性模型以支持形式化分析和推理.该模型具有一定程度的自管理能力,采用分级处理的方式应对各种程度的可信性威胁,满足了关键任务系统对可信性的特殊需求.在此基础上,从稳态概率角度提出了一种自律可信性度量方法.最后,结合具体实例对模型参数对自律可信性的影响进行了初步分析.实验结果表明,增大关键任务系统可信性威胁检测率和自恢复成功率,可在较大范围内提高系统的自律可信特性.
文摘The software architecture is a design of application system,performance evaluation of software architecture during the early stage of their development is really attractive. This paper proposes a new method for software architecture performance modeling. To achieve this ,it adds new calculus into stochastic process algebra(SPA in short),the developed SPA called extended stochastic process algebra (ESPA in short). By ESPA,performance evaluation and software architecture modeling can combine perfectly. It defines a few performance terms for software architecture using reward structure derived from ESPA. To explain the performance terms it also designs an experiment.
