Nowadays, Information and Communication Technologies industry must deal with the rapid growth of data. These data require a large storage capacity, processing power and considerable calculation, consuming a wide bandwidth. The increasing demand for resources, decrease the technological infrastructure life cycle. The investment in new infrastructure becomes risky and expensive. However, the development of the Internet enables infrastructure sharing. A very promising new paradigm is emerging. This technology is known as resources virtualization. It’s a new way of conceptualizing oriented virtual organizations and provides infrastructure as service. Based on resources sharing, virtualization will not only reduce costs but also increase yields of resources. Like servers that are often underutilized. Nevertheless, distributed and heterogeneous environment of virtual organizations makes the resource discovery and selection a difficult task.
In this master thesis, we propose a method for resource discovery and research, based on artificial intelligence using Bayesian networks and ontology. This method enables Web resources search in a wide range of resources representation and description. Further to its originality, our method has the merit of being generic. It supports different search criteria and optimization methods as well representations of resources. The evaluation of our approach has been undertaken as part of the Green Star Network (GSN) test bed. The tests have shown promising results.
| Date | 15 Mar 2011 |
|---|
| Original language | French |
|---|
| Awarding Institution | - École de technologie supérieure
|
|---|
| Supervisor | Mohamed Cheriet (Supervisor) |
|---|
Daouadji, A. (Author),
Cheriet (Supervisor),
15 Mar 2011Student thesis: Master's thesis › Master in Engineering: Automated Manufacturing Engineering