Search

Fixed networks play an increasingly important role in supporting broadband services to homes, offices, shopping centers, business buildings, factories, smart cities, and much more. Reaching closer to end-user access points in rooms, office desks, and even factory machinery, optical fiber will realize its full potential to support a fully connected, intelligent world with high bandwidth, high reliability, low latency, and low energy consumption. With the fiber-to-everywhere vision, the European Telecommunications Standards Institute (ETSI) established an industry specification group (ISG) dedicated to the definition and specification of the 5th generation fixed network (F5G) in 2020. In this article, we describe the overall architecture of F5G, which consists of three interacting planes, the management, control and analysis plane, the service plane, and the underlay plane. F5G enables the quality of service (QoS) of each of the various services carried to be satisfied via end-to-end (E2E) network slicing over the customer premises network, access network, aggregation network, and core network segments. With the comprehensive service-oriented features of F5G, 14 use cases have been conceived under three main application scenarios, enhanced fixed broadband, guaranteed reliable experience, and full fiber connection. We show that F5G is capable of supporting these use cases with the requested QoS in terms of bandwidth, latency, agile service creation and bandwidth adjustment, fine granularity of bandwidth reservation, and automated E2E network orchestration and management. To further show the capabilities of the F5G architecture, we discuss the E2E network slicing in a cloud virtual reality demonstration, as well as a time-sensitive optical network for supporting cloud-based industrial applications. Finally, future perspectives of F5G and its standardization are discussed.

International audience

The Internet Protocol (IP) has been proven very flexible, being able to accommodate all kinds of link technologies and supporting a broad range of applications. The basic principles of the original Internet architecture include end-to-end addressing, global routeability and a single namespace of IP addresses that unintentionally serves both as locators and host identifiers. The commercial success and widespread use of the Internet have lead to new requirements, which include internetworking over business boundaries, mobility and multi-homing in an untrusted environment. Our approach to satisfy these new requirements is to introduce a new internetworking layer, the node identity layer. Such a layer runs on top of the different versions of IP, but could also run directly on top of other kinds of network technologies, such as MPLS and 2G/3G PDP contexts. This approach enables connectivity across different communication technologies, supports mobility, multi-homing, and security from ground up. This paper describes the Node Identity Architecture in detail and discusses the experiences from implementing and running a prototype.

Service specific overlays are overlay networks build and setup for a single service. Since different services need different types of overlays the setup process is pretty different. Current approaches normal allow to send in a overlay specification into the overlay provisioning system (mainly centralized). In our approach, we use pattern-based management in order to run the process completely distributed but still having a basic communication pattern to be reused for other overlays with different requirement. In this talk, we present the concept and a first implementation on the Simpson pattern simulator of KTH.

From 03.04.05 to 06.04.05, the Dagstuhl Seminar 05142 ``Disruption Tolerant Networking'' was held in the International Conference and Research Center (IBFI), Schloss Dagstuhl. During the seminar, several participants presented their current research, and ongoing work and open problems were discussed. Abstracts of the presentations given during the seminar as well as abstracts of seminar results and ideas are put together in this paper. The first section describes the seminar topics and goals in general. Links to extended abstracts or full papers are provided, if available.

Disruption Tolerant Networking (DTN) is a new area of research to improve network communication when connectivity is periodic, intermittent, and/or prone to disruptions. A seminar on DTN was held at at Schlo�������� Dagstuhl, Germany, from 3 to 6 April 2005. Researchers from different fields discussed their approaches to dealing with delays, intermittent connectivity, and the potential non-existence of an end-to-end path in a number of different environments. The two major areas identified were: (1) dealing with delay and disruption in the present Internet in the context of wireless, mobile, and nomadic communications, supporting existing applications and (2) addressing new applications with a focus on exploiting discontinuous connectivity and opportunistic contacts for asynchronous communications. This article briefly reviews the seminar presentations and discussions.

Traditional traffic measurements meter throughput on time scales in the order of 5 minutes, e.g., using the Multi Router Traffic Grapher (MRTG) tool. The time scale on which users and machines perceive Quality of Service (QoS) is, obviously, orders of magnitudes smaller. One of many possible reasons for degradation of the perceived quality, is congestion on links along the path network packets traverse. In order to prevent quality degradation due to congestion, network links have to be dimensioned in such a way that they appropriately cater for traffic bursts on time scales similarly small to the time scale that determines perceived QoS. It is well-known that variability of link load on small time scales (e.g., 10 milliseconds) is larger than on large time scales (e.g., 5 minutes). Few quantitative figures are known, however, about the magnitude of the differences between fine and coarse-grained measurements. The novel aspect of this paper is that it quantifies the differences in measured link load on small and large time scales. The paper describes two case studies. One of the surprising results is that, even for a network with 2000 users, the difference between short-term and long-term average load can be more than 100%. This leads to the conclusion that, in order to prevent congestion, it may not be sufficient to use the 5 minute MRTG maximum and add a small safety margin.

Berichte aus der Informatik, ISBN:3-8265-7583-0

The paper addresses the question on how the benefits of active networking, such as customized packet processing inside the network and flexible service management, can be realized in a telecom environment. More precisely, we ask: How can a network provider, whose infrastructure is based on active networking technology, support a large number of customers, all of which independently install and run their own customized active services in the providers domain? Our approach is based on network virtualization, and the goal of this paper is to demonstrate that virtualization of active networks can be achieved with considerable benefits for customers and providers and with limited costs. Our work uses the concept of the Virtual Active Network (VAN), a generic service that is offered by an active network provider to customers. We have realized the VAN concept and implemented a VAN provisioning and management architecture on ANET, an active networking platform we have developed. ANET is an all-software, functional prototype of an active network, which allows for experimentation with great flexibility. Further, we have worked out a design for a VAN-enabled node operating system for a high-performance active network node which is currently being built by our laboratory at ETH Zurich in collaboration with a group at Washington University in St. Louis., TIK Report, 79