Context-awareness and demand-driven evaluation in heterogeneous distributed systems

本研究提案旨在界定一个理论上合理的框架， 上下文感知系统互操作性协议、上下文感知中间件以及基于需求驱动评估方法的编程语言语义的相关定义和实现。在这个框架中，我们将我们过去的理论和实践成果结合起来，包括：（1）一个可靠的上下文演算理论;（2）一个需求驱动的求值语义，包括一个实现这种实现所必需的语言无关的类型系统;（3）验证方法与可伸缩性和易用性，以改进特定应用程序和中间件领域（例如SOA）中的现有方法，以及云、网格和移动的计算，这些计算越来越具有上下文感知能力。 上下文感知及其相应的需求驱动计算模型可以应用于 面向服务的体系结构（SOA）是分布式计算的一种模型，它使面向服务的客户端能够收集自己的上下文，并根据自己的上下文提出面向服务的服务器需求，并根据服务需求的上下文定义服务。在另一个频谱上，诸如网格计算和云计算的新计算模型现在通常依赖于上下文的概念，以便约束或指导处理需求所使用的计算资源。同样，上下文感知中间件的概念，加上相应的需求驱动的评估模型，正好符合当前的需要。我们将在上述SOA、云网格和移动的计算及其应用领域中重用和扩展我们以前最好的理论成果及其实际的基于组件的实现，以统一方法，在统一的研究评估平台上测试和验证各种上下文感知范式的可扩展性和可用性。

This research proposal aims at the definition of a theoretically sound framework to introduce the notion of context-aware system interoperability protocol, context-aware middleware, as well as the associated definition and implementation of a programming language semantics based on a demand-driven evaluation method. We propose to integrate our past theoretical and practical results in this framework, which include: (1) a sound theory of context calculus; (2) a demand-driven evaluation semantics, including a language-independent type system necessary for such an implementation; (3) validation of the approach vs scalability and ease of use as to improve existing methodologies in the specific application and middleware domains, e.g. SOA, and cloud, grid, and mobile computing which are increasingly context-aware. Context awareness and its corresponding demand-driven computation model can be applied to the service-oriented architecture (SOA) model of distributed computing, enabling service-oriented clients to gather their own context and make service-oriented server demands based on their own context, and the services to be defined according to the context of the service demands. On another spectrum, new computing models such as grid computing and cloud computing are now generally relying on the notion of context in order to constraint or direct the computational resources being used by processing demands. Again, the notion of context-aware middleware, coupled with a corresponding demand-driven evaluation model fits exactly this current need. We will re-use and expand our previous best theoretical results and their practical component-based realization in the above mentioned domains of SOA, cloud grid, and mobile computing and their applications, to unify the approach, test and validate the scalability and usability of various context-aware paradigms in a uniform research evaluation platform.