Context-Oriented Programming: A Programming Paradigm for Autonomic Systems

Abstract

Dynamic software adaptability is one of the central features leveraged by autonomic computing. However, developing software that changes its behavior at run time adapting to the operational conditions is a challenging task. Several approaches have been proposed in the literature to attack this problem at different and complementary abstraction levels: software architecture, middleware, and programming level. We focus on the support that ad-hoc programming language constructs may provide to support dynamically adaptive behaviors. We introduce context-oriented programming languages and we present a framework that positions the supported paradigm in the MAPE-K autonomic loop. We discuss the advantages of using context-oriented programming languages instead of other mainstream approaches based on dynamic aspect oriented programming languages and present a case study that shows how the proposed programming style naturally fits dynamic adaptation requirements. Finally, we discuss some known problems and outline a number of open research challenges.