Previsão de Carga para Economia de Energia em Aglomerados de Servidores Web

Carlos Henrique Sant' AnaJ. C. B. LeiteDaniel Mossé

The complexity and requirements of web applications are increasingin order to meet more sophisticated business models (web services and cloudcomputing, for instance). For this reason, characteristics such as performance,scalability and security are addressed in web server cluster design. However,due to the energy crisis, the energy consumption in this type of environment becamea main concern. Aiming to avoid this adverse scenario, this article showshow energy reduction can be applied in a web server clustered environment. Theenergy reduction policy used in this paper uses load forecasting combined withDVFS (Dynamic Voltage and Frequency Scaling) and dynamic configuration techniques.To validate this predictive policy, a web application running a realworkload profile was submitted to a server cluster testbed. In addition a criteriafor quality of service was used to evaluate system's performance.

Caso o link acima esteja inválido, faça uma busca pelo texto completo na Web: Buscar na Web

Biblioteca Digital Brasileira de Computação - Contato:
     Mantida por: