Riccardo Melen

Optimal cost/performance design of ATM switches

This paper proposes a methodology for performing an evaluation and optimization of the cost of an ATM switching architecture under performance constraints given in terms of virtual connection blocking probability. An analysis of blocking networks is developed, and combined with known results concerning nonblocking networks, provides a theoretical model which relates traffic characteristics, network topology and blocking probability in a multirate/multiservice broadband environment. An analysis of the characteristics determining the cost of a generic ATM switch implementation follows. The model is oriented to optimize both the topological parameters and the speed advantage, with respect to the main cost factors of VLSI-based switching networks i.e., components count and complexity, interconnection costs. >

Evaluating the limit behavior of the ATM traffic within a network

This paper is concerned with the ATM traffic characterization within the network. Most of the work performed up to now has studied the effects of traffic on the access multiplexer and the first switch of an ATM network. Various source models were assumed to generate the ATM traffic. So, while the performance of a single switch node has been exhaustively examined, the statistical behavior of the traffic modified as it crosses the network has not been thoroughly analyzed yet. This paper, through an analysis of a network of cascaded queues, indicates that limit distributions exist in the statistical behavior of the traffic streams and in the queue performance, although a formal proof is believed to be very hard to obtain. The first modelling step consists of deriving the exact interdeparture time distribution for the cells of a reference-connection arriving to the output queue of a switch node with a general interarrival time distribution and multiplexed with a background traffic stream. The analysis is iterated through a long sequence of cascaded output queues, until the interdeparture time distribution converges. Simulations show that the analytical results are accurate at each stage of the network under the hypothesis of independent queues, and are also good approximations in the case of correlated queues. This study shows that the queue performance at the limit point is always better than the M/D/1 case. The distributions found in this way depend only on the connection bandwidth and on the background traffic behavior. The initial characteristics of a connection (burst length distributions and burst interarrival time distributions) only influence the convergence speed, not the limit distribution.

Performance and dimensioning of a hierarchical video storage network for interactive video services

This paper examines a network architecture designed to offer interactive video services to residential subscribers. Its main feature consists in the possibility of storing heavily requested material in video “cache” locations at the local switch level. Various policies for the management of the resulting hierarchical storage system are defined and their effectiveness, in terms of success ratio of the local accesses, is evaluated via simulation. A simple evaluation of the cost trade-off between distributing video programs in local memories and keeping them in a centralized video server is carried out as well, considering both leased line and switched line solutions for long distance transmission. The difference due to the non stationary behaviour of the user requests is also evaluated.

The role of local storage in supporting video retrieval services on ATM networks

This paper studies a reference network architecture for video information retrieval, based on hierarchical storage of the video segments and asynchronous transfer mode (ATM) transport capability on the long distance network. The storage associated to local switching nodes can be exploited in two complementary ways, either for caching the most requested video programs, and for decoupling the transmission speed on the long distance network from the natural playout speed of the video segments. We show that both approaches lead to a substantial increase of the load carried on the long distance network with a given service requirement. Moreover, we show that a greater efficiency is obtained when a suitable integration of the two approaches is considered.

An Experimental Open Architecture to Support Multimedia Services Based on CORBA, Java and WWW Technologies

The convergence of interactive multimedia services and Internet services characterises the new services scenario and also highlights the important role of the terminals and their intelligent capabilities. Thus the scenario moves from the classical telecom setting where the “intelligence” is concentrated and embedded into the network (switches) towards a computer networking scenario in which more “intelligence” is located at the periphery (hosts, computers and terminals). The primary objective of the ORCHESTRA project was to demonstrate by fast prototyping an experimental control architecture capable of complying with a large number of the emerging requirements for the support of advanced services in broadband networks, with particular emphasis on emerging technology integration and interoperability.

ORCHESTRA: an experimental agent-based service control architecture for broadband multimedia networks

This paper deals with the development of a prototype software architecture, named ORCHESTRA, aimed at the provision of advanced multimedia services in broadband networks, and supporting by both Internet and ATM signaling protocols. The focus is an experimental service control architecture capable of fulfilling the major requirements of emerging broadband multimedia services. The key points of the implementation are: (1) the use of software agents to enhance the architectures flexibility, autonomy and distribution, and (2) the use of global service brokers for orchestrating services among terminals, servers and network elements, and for the control and management of network resources.

Non stationary request distribution in video on demand networks

Several works have recently appeared in the literature on the design, performance and dimensioning of networks supporting video on demand and multimedia services. It has been shown that performance is strongly dependent on the model adopted to describe the users behaviour in terms of requests for programs. In this paper we provide a detailed investigation of a non stationary model for the users request distribution. Numerical results are provided for a network scenario proposed in the literature to investigate the error achieved when the non stationary distribution is approximated with a stationary exponential one.

Effects of multiple node crossing on ATM traffic performance

It is commonly recognized that traffic statistics change when a stream of ATM cells crosses several network nodes. In particular, some cells may experience a shorter delay than the previous ones, and thus the peak rate of the traffic stream may increase. In order to control this “cell clumping” effect, on one side spacing techniques have been proposed to restore the original peak rate of the sources at the outlets of each network node, while on the other side bounds to keep into account the peak rate increase have been suggested. Although some contributes have evaluated the effects of multiple node crossing on the distribution of the cell interarrival time, little attention has been paid to the effect of these modifications in terms of cell loss probability. Scope of this paper is to show, by adopting a realistic simulation model, that multiple node crossing produces a negligible effect on cell loss probability, when the admitted traffic is suitably regulated by a connection admission control rule. Moreover, a notable result is that, in the case of large buffers at the network nodes and traffic loads well over the limits imposed by the CAC rule, the per node cell loss probability eventually reduces as network nodes are crossed.

Testing methods and equipments for ATM switching nodes

After several years of theoretical studies and laboratory experiments, the ATM tech- nique has now reached a sufficient degree of maturity that makes it ready to be experimented in the field. ATM substantially differs from traditional switching and transmission techniques for the statistical multiplexing introduced by the asynchronous technique. It turns out that the testing pro- cess is a formidable challenge for the operator, due to both the functional complexity foreseen for the B-ISDN switching systems and to the sophisticated performance issues involved. This paper is devoted to a first assessment of the relatively unexplored problem of specifying ATM switching equipment and testing its conformance to the specification. It begins by reviewing some concepts regarding the testing of existing switching systems, then examines the peculiarities of ATM, out- lines the main critical issues and their possible solution.

Free-space architecture for an ATM Header processing function

Header Error Control in all-optical ATM switching nodes is discussed. An architecture of an error detection subsystem is designed suitable for free-space parallel optical implementation.