Addressing Scheduling Issues in Context Information Management
Maria Chantzara, Nikolaos Vardalachos, Μιλτιάδης Αναγνώστου · 2006
The provision of context-aware services, that is crucial for mobile and ubiquitous computing, prerequisites appropriate management methods for acquiring, processing and distributing context information. This paper focuses on the challenge that refers to the seamless provision of real-time access to the dynamically changing context data. It studies the problem of handling the order of the information update requests issued by the context sources. The objective is to maximize the freshness of the delivered data and satisfy the timing constraints of the context consumers, when firm deadlines, non-pre-emptive order and limited processing resources are considered. In order to tackle the conflicting requirements, the Importance / Popularity / Urgency- Aware Scheduling Algorithm is developed. The proposed algorithm is evaluated through extensive performance studies.