1,721,071 research outputs found

    Rete e servizi: tra criticità e scenari futuri

    Full text link
    Nel periodo di lockdown, che abbiamo di recente vissuto, le tecnologie digitali sono state fondamentali per il lavoro da casa, per lo svago, per la comunicazione e la condivisione delle esperienze, di fatto per aiutarci ad andare oltre il lockdown stesso. Queste tecnologie, già molto presenti nella nostra vita, hanno assunto un ruolo così fondamentale in modo improvviso e senza alcuna pianificazione preventiva, sia a livello tecnico sia a livello gestionale. Il fatto che le cose abbiano funzionato, è già di per sé un risultato notevole, che merita un’analisi attenta. In questo articolo proponiamo vari spunti di analisi, chiarendo innanzitutto l’importante distinzione fra rete e applicazioni. Sottolineeremo come queste due componenti rappresentino ecosistemi distinti che obbediscono a regole e dinamiche disgiunte e come sia necessario prevedere in futuro chiare politiche capaci di governare in modo chiaro gli obiettivi per cui queste tecnologie vengono dispiegate quando diventano centrali nella vita dei cittadini, cosa non avvenuta appieno fino ad oggi.During the recent lockdown period, digital technologies have played a fundamental role. They freed us from a significant part of the lockdown constraints, by enabling smart working, keeping social connections alive, and allowing us to share our feelings and experiences. These technologies were already pervasive, but the pandemic crisis elevated them to a key factor in our lives in an instant, with no prior technical or management planning. The simple fact that in the end everything just seemed to work is a non-trivial outcome. It is worth a detailed analysis. In this paper, we offer some leads, starting from the important distinction between network infrastructure and services. We highlight that these two components represent ecosystems ruled by different dynamics. We claim that, for the future, well-defined goals should be set for the adoption of these technologies, which play such a central role in citizens’ lives, and effective policies are needed to keep their deployment aligned with said goals, unlike what has happened until now

    Introduction

    No full text
    This book provides a throughout overview of IP over WDM technologies, as seen by a group of experts participating in the e-Photon/One+ and BONE networks of excellence of the VIth and VIIth framework programmes of the European Union. Both networks of excellence aimed at the integration of research in optical networks at the paneuropean level, with the creation of virtual centres of excellence in optical networks, technologies and services. The working groups on Optical Core networks gathered about a hundred of researchers from more than 20 Universities and research institutes in Europe. From this community stemmed the idea of merging the various expertises in a book reporting the most advanced viewpoints on this broad topic

    Optical Packet Switching

    No full text
    This chapter provides an outline of the main concepts and issues related to optical packet switching. The basic network functions required by this paradigm are discussed and references to past and current research on the topic are provided. Optical packet switching is a transport technique that assumes information to be organized in packets formed by a payload and an header both encoded as optical signals. The payload is transferred through the network without any optical to electronic conversion. The header is processed in the early phase in the electronic domain. Optical packet switching may be considered as a long-term and more flexible alternative to the circuit-switched optical networks currently being deployed by operators. This innovative paradigm aims at optimizing the utilization of the DWDM channels by means of fast and highly dynamic resource allocation, overcoming the inefficiency typical of the circuit transfer modes. Traditionally, packet transfer modes have proved to be very flexible by nature, with respect to bandwidth exploitation. In fact, link capacity is shared in time by means of statistical multiplexing, while contentions occurring at each node are solved by storing packets as long as the required resources become available again. Therefore, network links are in general used more efficiently than in circuit-switching

    Analytical model for anycast service provisioning in data center interconnections

    No full text
    This paper presents an analytical model to evaluate the performance of an optical wavelength-routed network for inter- and intra-data center interconnection, designed to work according to the cloud computing paradigm. Such a scenario calls for anycast routing of service requests to find a network path that best suits both connectivity and IT resource requirements. The analytical model is based on conventional queuing theory for loss systems combined with an ad-hoc combinatorial analysis of the anycast service alternatives. The model presented here is meant to provide engineering guidelines for data center interconnection networks supporting the cloud paradigm, in particular in terms of performance and cost of resource distribution

    Analysis of an ATM Shared Buffer Switch Loaded with Bursty Traffic

    No full text
    In this paper a modeling approach to performance evaluation of a shared buffer switching element is described, based on the well-known fluid model of producers and consumers (PC fluid model). A procedure is outlined that leads to a suitable characterization of some typical parameters of the producer and consumer fluid model, making it representative of the shared buffer switch. Simulation analysis is used to investigate the relationships between the behavior of the shared buffer switching element and of the PC fluid model. In the paper it is shown that by means of a suitable fitting of one parameter characterizing the PC fluid model, it is possible to make it representative of the shared buffer in the region of interest for ATM applications. This in spite of the actual operating differences between the real system and the PC fluid model. Numerical results regarding cell loss probability performance and dimensioning of 4 by 4 and 8 by 8 switches are presented and suitably discussed

    A Wireless ATM LAN with an 802.11-like Data Link Layer

    No full text
    In this paper a semi-analytical procedure isemployed to evaluate the performance of a wireless ATMlocal area network based on the IEEE 802.11 MAC protocol,taking into account the effects of propagation over a real radiochannel. Numerical results are presented as a function of theMAC frame length and of the signal-to-noise ratio in order toevaluate network performance and to provide guidelines forthe overall system design

    The Network of Excellence e-Photon/ONe

    No full text
    e-Photon/ONe is a Network of Excellence (NoE) on optical networks funded by the EU in the VI Framework Program. NoEs aim at fostering the integration of the activities of the involved researchers and institutions, exploiting complementarity to achieve critical mass. “Virtual Departments” in e-Photon/ONe coordinate research activities on specific thematic areas. In this paper we present a summary of Virtual Departments views on the state-of-the-art and on future trends and key issues in research on optical networks

    Network resource allocation in data center interconnection with anycast service provisioning

    No full text
    The joint management of IT and network resources is a key aspect for dynamic inter-data center interconnection networks designed to work according to the cloud computing paradigm. Such a scenario calls for the anycast routing of service requests to find a data channel that best suits both connectivity and IT resource requirements. This paper presents different network resource allocation and release policies across an inter-data center interconnection network aimed at a balanced accommodation of network resources to improve performance. Such policies are evaluated and compared in terms of service set-up blocking probability considering also configuration latency of real network devices that play a significant role along with network bandwidth availability across the network. Results show that the best choice is to release network resources as soon as possible, even though this means increasing the service blocking due to network reconfiguration latency
    corecore