No abstract

Self-adaptive enterprise applications have the ability to continuously reconfigure themselves according to changes in their execution contexts or user requirements. The infrastructure managing such systems is based on IBM's MAPE-K reference model: a Monitor and an Analyzer to sense and interpret context data, a Planner and an Executor to create and apply structural adaptation plans, and a Knowledge manager to share relevant information. In this paper we present a formal model, built on the principles of constraint satisfaction, to address dynamic adaptation planning for self-adaptive enterprise applications. We formalize, modify and extend the approach presented in [H. Arboleda, J. F. Díaz, V. Vargas, and J.-C. Royer, "Automated reasoning for derivation of modeldriven spls," in SPLC'10 MAPLE'10, 2010, pp. 181-188] for working with self-adaptation infrastructures in order to provide automated reasoning on the dynamic creation of structural adaptation plans. We use a running example to demonstrate the applicability of such model, even in situations where complex interactions arise between context elements and the target self-adaptive enterprise application. © 2016 The Authors

Companies’ operations have become over-dependent on their supporting enterprise software applications. This situation has placed a heavy burden onto software maintenance teams who are expected to keep these applications up and running optimally in varying execution conditions. However, this high human intervention drives up the overall costs of software ownership. In addition, the current dynamic nature of enterprise applications constitutes challenges with respect to their architectural design and development, and the guarantee of the agreed quality requirements at runtime. Efficiently and effectively achieving the adaptation of enterprise applications requires an autonomic solution. In this paper we present SHIFT, a framework that provides (i) facilities and mechanisms for managing self-adaptive enterprise applications through the use of an autonomic infrastructure, and (ii) automated derivation of self-adaptive enterprise applications and their respective monitoring infrastructure. Along with the framework, our work leads us to propose a reference specification and architectural design for implementing self-adaptation autonomic infrastructures. We developed a reference implementation of SHIFT; our contribution includes the development of monitoring infrastructures, and dynamic adaptation planning and automated derivation strategies.