The second generation of Web sites provides more complex services than those related to Web publishing. Many users already rely on the Web for up-to-date personal and business information and transactions. This success motivates the need to design and implement Web architectures being able to guarantee the service level agreement that will rule the relationship between users and Web service providers. As many components of the Web infrastructure are beyond the control of Web system

more »

... s, they should augment satisfaction percentage of the assessed service levels by relying on two mechanisms that can be integrated: differentiated classes of services/users, Web systems with multi-node architectures. The focus of this paper is on this latter approach. We review systems where replicated Web services are provided by locally and geographically distributed Web architectures. We consider different categories of Web applications, and evaluate how static, dynamic and secure requests affect performance and quality of service of distributed Web sites. of the assessed service levels, they can rely on two classes of actions that are not mutually exclusive: Differentiated Web services. It requires the definition of classes of users/services, choice of the number of priority levels, guarantee of different SLAs through priority dispatching disciplines [6, 15, 20] and monitors for starvation of low priority services. Architecture design. The goal is to find the right architecture that guarantees the SLA on all Web users/services. The three directions are: scale-up by adding memory and CPU power to the single server, local scale-out by replicating servers in a local area, global scale-out by replicating servers in a geographical context. The focus of this paper is on the architecture design, while we leave to future work the combination of the two 0-7695-