Empowering Grids with Flexibility to Cope with Uncertainties

Abstract

Grids are systems that involve coordinate resource sharing and problem solving in heterogeneous dynamic environments. In order to make grids effective it is necessary to provide them with the capability of dealing with uncertainties on resource availability and estimation of applications computational and communications demands. This paper presents a procedure for self-adjusting the resources allocated to an application and a scheduler which takes into consideration the uncertainties on the estimation of applications demands. Moreover, a brief survey of resource allocation schemes in different existing grids is provided.