SGER: A Model for Highly Dynamic Real-Time Systems - Coordination+Timing Constraint+Functionality

移动设备和无线网络技术的快速发展以及嵌入式设备的日益普及扩展了传统实时系统的应用范围。这种系统现在包括嵌入式组件之间的协调和组件之间通信拓扑的动态重新配置。这种高动态的嵌入式混合控制系统存在许多挑战。传统上，编程语言社区关注于分布式组件之间交互语义的建模和推理。与此同时，实时计算界关注的是如何管理CPU和网络通信资源，使实时任务可以预见地满足端到端的时间约束。本研究试图通过抽象动态实时系统中的正交关注点，然后使用这些单独指定的组件来支持运行时集成系统行为，从而将这两个领域连接起来。这项研究正在开发一个框架来解决异步动态实时嵌入式系统中的关键问题：协调、实时约束和可重构性，以模块化的方式允许关注点分离。框架的原型实现包括用于指定独立于单个计算对象的内部行为的实时约束和协调需求的定义语言构造。它还提供了组合组件以实现集成系统需求的运行时支持。特别是，时序约束的语言抽象用于支持运行时调度器，而交互拓扑的语言抽象用于支持消息路由。这样的运行时框架不仅可以用作测试平台来验证所建议的模型，还可以验证方法的实际方面。研究技术的成功成果将对未来异步分布式实时嵌入式系统的开发产生深远的影响。实时嵌入式系统协调基板的强大之处在于它被设计成模块化和动态可重构的，以适应分布式实时嵌入式系统的动态特性。模块化和关注点分离不仅提高了软件在计算领域和协调领域的可重用性，而且简化了大型复杂实时嵌入式系统的设计、开发和分析，从而提高了系统的可靠性。

The rapid advancement of mobile devices and wireless networking technologies and increasing use of embedded devices has extended the scope of traditional real-time systems. Such systems now include coordination among embedded components and dynamic reconfiguration of communication topologies among components. There are many challenges in such highly dynamic embedded hybrid control systems. Traditionally, the programming language community focused on modeling and reasoning about the semantics of interactions between distributed components. Meanwhile the real-time computing community focused on how to manage CPU and network communication resources so that real-time tasks can predictably meet their end-to-end timing constraints. This research seeks to bridge the two areas by abstracting orthogonal concerns in dynamic real-time systems, and then using these separately specified components to support runtime-integrated system behaviors.This research is developing a framework to address key issues in asynchronous dynamicreal-time embedded systems: coordination, real-time constraints, and re-configurability in a modular fashion that permits separation of concerns. A prototype implementation of the framework includes the defining language constructs for specifying real-time constraints and coordination requirements independent of the internal behavior of individual computational objects. It also provides runtime support that combines the components to achieve integrated system requirements. In particular, the language abstraction for timing constraints is used to the support run-time scheduler, while the language abstraction for interaction topology is used to support message routing. Such a runtime framework cannot only be used as a test-bed to validate the proposed model, but also can verify the practical aspects of the approach.Successful results of the research techniques are expected to have a profound impact on the ease of development of future asynchronous distributed real-time embedded systems. The power of the coordination substrate for real-time embedded systems lies in the fact that it is designed to be modular and dynamically reconfigurable to suit the dynamic nature of distributed real-time embedded systems. The modularity and separation of concerns not only increases the software reusability in both computational domain and coordination domain, but also simplifies the design, development and analysis of large complex real-time embedded systems and hereby increases system dependability.