The main objective of this sequel is to solve the out-of-sequence problem that occurs in the load balanced Birkhoff-von Neumann switch with one-stage buffering. We do this by adding a load-balancing buffer in front of the first stage and a resequencing-and-output buffer after the second stage. Moreover, packets are distributed at the first stage according to their flows, instead of their arrival times in Part I. In this paper, we consider multicasting ows with two types of scheduling policies: the First Come First Served (FCFS) policy and the Earliest Deadline First (EDF) policy. The FCFS policy requires a jitter control mechanism in front of the second stage to ensure proper ordering of the traffic entering the second stage. For the EDF scheme, there is no need for jitter control. It uses the departure times of the corresponding FCFS output-buffered switch as deadlines and schedules packets according to their deadlines. For both policies, we show that the end-to-end delay through our multistage switch is bounded above by the sum of the delay from the corresponding FCFS output-buffered switch and a constant that only depends on the size of the switch and the number of multicasting flows supported by the switch.

Distributed content-based publish-subscribe middleware provides the necessary decoupling, flexibility, expressiveness, and scalability required by modern distributed applications. Unfortunately, this kind of middleware usually does not provide reliability guarantees, as this problem has been thus far largely disregarded by the research community and solutions developed in other contexts are not immediately applicable.

Secure group communication has applications in multimedia conferencing, stock quote distribution, shared work space, distributed interactive simulation, etc. Some of these applications engage in one to many communication while others involve many to many communication. Developing scalable secure multicasting protocols is a challenge, particularly when multiple senders are present. Several protocols have been proposed to support secure group communication involving a single sender. In this paper we classify and describe existing secure one to many multicast protocols. Our survey is augmented by a numerical comparison of hierarchical secure group communication protocols, using Mbone group membership traces. We also describe the few secure many to many protocols proposed in the literature. 1 Introduction With the widespread use of the Internet, securing data transmissions is an important requirement for many applications. Secure group communication has several applications in mu...

Abstract- The simplicity of multicast as a communication primitive belies its broad utility as a building block for distributed applications. Nevertheless, creating and maintaining multicast structures can be challenging, particularly when networks are transient and/or dynamic. We introduce a new unstructured multi-source multicast (UMM) overlay approach that we argue is less complex than, but as efficient as, current state-of-the-art solutions based either on structured overlays or on running full routing protocols at the overlay level. UMM builds a base overlay independently from the routing mechanisms employed to route messages. On top of this base overlay, it selects distribution trees for each multicast source by first flooding the base overlay and then using the implicit information contained in duplicated messages to select and filter out redundant tunnels. Simple heuristics are used to maintain and evolve both the base overlay and the multicast distribution trees in response to changes in the set of overlay participants or in underlying network conditions. We experiment on a 65-node PlanetLab deployment and on ModelNet emulated distributed platforms to quantify the overheads associated with UMM operation and to explore its performance and adaptability to changes in the underlying network conditions. I.

Distributed content-based publish-subscribe middleware provides the necessary decoupling, flexibility, expressiveness, and scalability required by modern distributed applications. Unfortunately, this middleware usually does not provide reliability, especially in the presence of highly reconfigurable scenarios. Indeed, this problem has been thus far largely disregarded by the research community and solutions developed in other contexts are not immediately applicable.

We would like to thank our advisors Prof. Gian Pietro Picco and Prof. Gianpaolo Cugola for trusting, encouraging, and supporting us all throughout our thesis work. Through advice and lively discussions, they have contributed a lot to the development of this work. We also want to thank Prof. Carlo Ghezzi for arousing our interest in Software Engineering during his lectures and for his support in many occasions. We developed our research at the Software Engineering Laboratory of Politecnico di Milano. In this environment we had the opportunity to meet people that in so many ways contributed to this work. In addition to the ones we already mentioned, we would like to express our gratitude to Davide Balzarotti, Claudio Casoli, Mirco Cesarini, Davide Frey, Vincenzo Martena, Roberto Tedesco, Luca Zanolin and many others. We are especially grateful to our class mates for their friendship and support