The advent of the mobileWeb and the increasing demand for personalized contents arise the need for computationally expensive services, such as dynamic generation and on-the- fly adaptation of contents. Providing these services exacer- bates the performance issues that have to be addressed by the underlying Web architecture. When performance issues are addressed through geographically distributed Web sys- tems with a large number of nodes located on the network edge, the dispatching mechanism that distributes requests among the system nodes becomes a critical element.

In this paper, we investigate how the granularity of re- quest dispatching may affect the performance of a dis- tributed Web system for personalized contents. Through a real prototype, we compare dispatching mechanisms oper- ating at various levels of granularity for different workload and network scenarios. We demonstrate that the choice of the best granularity for request dispatching strongly de- pends on the characteristics of the workload in terms of heterogeneity and computational requirements. A coarse- grain dispatching is preferable only when the requests have similar computational requirements. In all other instances of skewed workloads, that we can consider more realistic, a fine-grain dispatching augments the control on the node load and allows the system to achieve better performance.