QINNA: A COMPONENT-BASED FRAMEWORK FOR RUNTIME SAFE RESOURCE ADAPTATION OF EMBEDDED SYSTEMS
Laure Gonnord, Jean‐Philippe Babau · 2009
Abstra t. Even if hardware improvements have in reased the performan e of embedded systems in the last years, resour e problems are still a ute. The persisting problem is the onstantly growing omplexity of systems, whi h in rease the need for reusable developement framework and pie es of ode. In the ase of PDAs and smartphones, in addition to lassi al needs (safety, se urity), developers must deal with quality of servi e (QoS) onstraints, su h as resour e management. Qinna was designed to fa e with these problems. In this paper, we propose a omplete framework to express ressour e onstraints during the developpement pro ess. We propose a omponent-based ar hite ture, whi h generi omponents and algorithms, and a developpement methodology, to manage QoS issues while developing an embedded software. The obtained software is then able to automati ally adapt its behaviour to the physi al resour es, thanks to degraded modes. We illustrate the methodology and the use of Qinna within a ase study. Key words: omponent, software ar hite ture, resour e dynami management, ase study. 1. Introdu tion. When fa ed to the problem of designing handled embedded systems, the developer must be aware of the management of limited physi al resour es (CPU, Memory). In order to develop multimedia software on su h systems where the quality of the resour e (network, battery) an vary during use, the developer needs tools to: • easily add/remove fun tionality (servi es) during ompilation or at runtime;