Specification and Modelling of Fault Tolerant and Self Healing Systems Visiting Fellowship for Professor Tom Maibaum

容错和自愈系统规范和建模 Tom Maibaum 教授访问奖学金

• 批准号: EP/D052912/1
• 负责人: Jeffrey Magee
• 金额: $ 4.07万
• 依托单位: Imperial College London
• 依托单位国家: 英国
• 项目类别: Research Grant
• 财政年份: 2006
• 资助国家: 英国
• 起止时间: 2006 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=EP%2FD052912%2F1
• 关键词: Specification; Modelling; Fault; Tolerant; Self

A system is said to be self-healing if it can recover from failures without external intervention. In other words, the system is capable of automatically re-organising itself to continue operating after part of it has failed. This is obviously closely related to the notion of fault-tolerance in which a system can operate normally despite experiencing failures. We use the term self-healing to mean a wider class of systems and degree of re-organization than is usually denoted by the term fault-tolerant.This proposal is to enable the Distributed Software Engineering (DSE) group at Imperial College to learn about and apply the description and reasoning techniques developed by Professor Tom Maibaum. In particular, Professor Maibaum is an international expert in modal action logics. These logics can allow notions of good behaviour and abnormal behaviour to be formally specified and as a result permit precise reasoning about fault tolerance. We intend to look at their application in self-healing systems which can dynamically reconfigure in response to changes in their environment.DSE has an existing toolset called the Labelled Transition System Analyser (LTSA) which permits the behaviour of a set of systems components to be specified using a process algebra notion. In addition, a composition of components can be mechanically checked to show that they satisfy certain precisely specified properties. However, currently, while some notions of self-healing can be expressed in the LTSA, this is cumbersome and the properties that can be checked not obviously or directly related to self-healing. The objectives of this project are thus to use Professor Maibaum's work to allow the succinct specification of both self-healing systems and the properties that they must satisfy.

如果一个系统可以在没有外部干预的情况下从故障中恢复，那么它就可以说是自我修复的。换句话说，系统能够在部分故障后自动重新组织自己继续运行。这显然与容错的概念密切相关，在容错中，系统即使遇到故障也可以正常运行。我们使用的术语自愈意味着更广泛的一类系统和程度的重组比通常表示的术语fault-tolerated.This建议是使分布式软件工程（DSE）集团在帝国理工学院学习和应用的描述和推理技术开发的教授汤姆Maibaum。特别是，Maibaum教授是模态动作逻辑的国际专家。这些逻辑可以允许的概念，良好的行为和异常的行为被正式指定，并作为一个结果，允许精确的推理容错。我们打算看看他们的应用程序中的自我修复系统，可以动态地重新配置，以响应其环境中的变化。DSE有一个现有的工具集，称为标记的过渡系统分析器（LTSA），它允许一组系统组件的行为被指定使用一个进程代数概念。此外，还可以对部件的组成进行机械检查，以表明它们满足某些精确指定的性能。然而，目前，虽然一些自我修复的概念可以在LTSA中表达，但这是繁琐的，并且可以检查的属性与自我修复没有明显或直接的关系。因此，本项目的目标是使用Maibaum教授的工作，以允许两个自我修复系统和它们必须满足的属性的简洁规范。