The Handbook of Engineering Self-Aware and Self-Expressive Systems

During previous Task 2.1, we have developed the notions of self-expression and different levels of computational self-awareness, inspired by corresponding psychological levels. In the context of architecture, we refer to the self-expression and different levels of computational self-awareness as capability of the systems to obtain and react upon certain knowledge. In this report, we study the cate-gorisation of different capabilities from the architecture perspective; this could create the possibility to ensure that, when designing self-aware systems, only relevant capabilities are included, and their inclusion justified by identified benefits. There is no need for a system to become unnecessarily complex, learning and maintaining capabilities which do nothing to advance the outcomes for that system, generating only overhead. We have codified the knowledge about how to architecture self-aware applications in the form of architecture patterns, each contains different capabilities. In this task, an architecture pattern refers to an architectural problem-solution pair using the capabilities in a given context. We have elicited some patterns, where each pattern is decentralised by design. That is, structurally our self-aware patterns resemble a peer-to-peer network of interconnecting self-aware nodes, varying only in the number of the capabilities and the type of interconnection between them. Until recently, architecture patterns for self-adaptive systems have received little attention [19]. Many existing patterns target specific application domains [15], limiting their reuse outside the domains where they were originally conceived. Weyns et al. [19] argued that UML notations are limited in their ability to characterise self-adaptive architecture patterns, hence they proposed a simple , generic notation for describing patterns for Monitor-Analyse-Plan-Execute (MAPE) architecture style. Our patterns are distinct in focus from Weyns' in the sense that while we model self-aware capability and knowledge concerns in the architecture, their attention was about MAPE component interaction. We adopt a pattern notation, similar to the one in [19] for describing our 4 Chen et al. self-aware patterns. Firstly, Weyns's notation [19] is simple and easy to comprehend. Secondly, we believe describing our self-aware patterns using existing notation in the self-adaptive community makes our work accessible to other researchers and paves the way for others to build on our work. Existing work on architecture patterns focus on modelling the components and connectors of architecture; in such context, components are specialisations of modules in the architecture and therefore have attributes and operations, but are also associated with the provide and required interfaces; and connectors could be the …