Abstract: An interesting dynamic process taking place in networks is the spontaneous spreading of information via rumor-like mechanisms. In addition to its relevance to social sciences, such mechanisms also form the basis of an important class of data dissemination protocols in computer and communication networks. These protocols, generally known as epidemic-style or gossip protocols, have recently gained prominence as a robust and scalable methodology for group communications in large distributed systems. In this talk I will describe the results of our Monte Carlo simulation studies of the impact of network architecture on the dynamics of gossip protocols in computer networks with scale-free and random graph connection topologies. Our studies provide new insights in the complex (and rather unexpected ) interplay between the gossip process and the underlying network topology. An analytical model of gossip dynamics in these networks will be presented.