Skip to Main Content
Whatever the circumstances, a complete transparency and a quick response time are expected by the users of a grid even when failures occur. To fulfill these two requirements, adaptive control mechanisms have to be proposed, on one hand to cope efficiently with the dynamic changes of the computing capacity of the grid (even if these changes are unpredictable) and, on the other hand to distribute the tasks among the resources in an efficient way (dynamic load balancing). In a hierarchical grid, both issues can be addressed using a common set of agreement services built on top of a consensus building block. The major aim of this paper is to investigate the benefits and the limits of this approach through some performance evaluations.