Analysis of large-scale distributed information systems

In this paper we study the effects of correlations between the interarrival times of different sen ice classes. An analysis sis of distributed information systems reveals that such interclass correlations exist, in part as a result of the interactions between the server and its clients. To gain insight into the performance implications of these correlations, we formulate a general stochastic model that explicit captures client-server interactions, and we derive a matrix-analysis of a specific instance of the model. Our results illustrate and quantify the impact that such inter-class correlations carl have on system performance.