A QoS-Based approach for dynamic web service composition.

Detalhes bibliográficos
Ano de defesa: 2009
Autor(a) principal: Frederico Guilherme Álvares de Oliveira Júnior
Orientador(a): Não Informado pela instituição
Banca de defesa: Não Informado pela instituição
Tipo de documento: Dissertação
Tipo de acesso: Acesso aberto
Idioma: eng
Instituição de defesa: Instituto Tecnológico de Aeronáutica
Programa de Pós-Graduação: Não Informado pela instituição
Departamento: Não Informado pela instituição
País: Não Informado pela instituição
Palavras-chave em Português:
Link de acesso: http://www.bd.bibl.ita.br/tde_busca/arquivo.php?codArquivo=854
Resumo: Web services have become a de facto standard for integration of systems in distributed environments by providing means for exchanging data in a standardized way over the Internet. In fact, this allows computer-computer interaction regardless the programming languages, platform and operating systems used. However, the way services are defined is not enough for computers to reason about the meaning of the information contained in that exchanged data. In the same way, information produced and consumed by Web Services can also be described so that computers can automatically discovery them in response to given request. This allows computers to dynamically combine several services when only one service is not able to provide the required functionality. However, there are scenarios where only the functionality may not be enough to fulfill the user requirements and a minimum level of quality, which is generally expressed by means of non-functional properties, should be guaranteed. In this thesis, an approach for dynamic web service composition that takes into account the composition overall quality is presented. The proposed approach relies on a heuristics to guide the composition in a efficient manner. In order to show the practical feasibility of the proposed approach, a composer prototype was developed and experimented with public service repository test sets. The results have shown the feasibility of the proposed approach even for large repositories as well as a significant improvement in the quality of the compositions found by this approach.