Skip to main content

Über dieses Buch

Architecture and Design for the Future Internet addresses the Networks of the Future and the Future Internet, focusing on networks aspects, offering both technical and non-technical perspectives. It presents the main findings of 4WARD (Architecture and Design for the Future Internet), a European Integrated Project within Framework Programme 7, which addressed this area from an innovative approach. Today’s network architectures are stifling innovation, restricting it mostly to the application level, while the need for structural change is increasingly evident. The absence of adequate facilities to design, optimise and interoperate new networks currently forces a convergence to an architecture that is suboptimal for many applications and that cannot support innovations within itself, the Internet. 4WARD overcomes this impasse through a set of radical architectural approaches, built on a strong mobile and wireless background. The main topics addressed by the book are: the improved ability to design inter-operable and complementary families of network architectures; the enabled co-existence of multiple networks on common platforms through carrier-grade virtualisation for networking resources; the enhanced utility of networks by making them self-managing; the increased robustness and efficiency of networks by leveraging diversity; and the improved application support by a new information-centric paradigm in place of the old host-centric approach. These solutions embrace the full range of technologies, from fibre backbones to wireless and sensor networks.



Chapter 1. Introduction

It starts by addressing some of the problems with current Internet networks, its core architecture and its evolution model. Current architectures are typically developed around layered models, and deficiencies have been shown, such as lack of support for QoS and seamless mobility, security vulnerabilities, and address shortage. The various forms of unwanted traffic, including spam, distributed denial of service, and phishing, are arguably some of the biggest problems. Changing business models are likely to have impact on the network. Privacy and accountability are other major issues. Next, a brief description of the 4WARD project is provided. It is followed by an overview of the current research and development activities being held in Europe, USA and Japan. It ends with a global view of the contents of the book.
Luis M. Correia, Henrik Abramowicz, Martin Johnsson, Klaus Wünstel

Chapter 2. A System Overview

The 4WARD System Model is described, defining the structure and behavior of a communication system that is to be constructed as well as its generativity, i.e., how bigger and more complex future systems and networks can be built by using a small set of generic concepts. It presents the project four tenets. Then, an Architecture Framework is shown, providing a unified component-based design process, which defines a seamless step-wise though iterative process for deriving a software-based network architecture using as input a set of technical requirements. The Architecture Pillars, described in detail, are: In-Network Domain Management, Network of Information, Generic Path, and the Physical Virtualized Substrate. The Architecture Framework is presented in terms of Strata, Netlets, and the Design Repository. The Design Process is also addressed.
Martin Johnsson

Chapter 3. Socio-economic

Non-technical drivers are addressed, when moving from the R&D stage to the real deployment of the technical and architectural innovations, grouped as: usage and services, socio-economic aspects, and regulation. Non-technical requirements are established, listed as twelve guidelines, which will have implications on network design rules. Four different scenarios are defined and developed, covering major aspects of both technical and non-technical areas: “Looking back from 2020: What made the old Internet break?”, “Novel applications that are not possible with the current Internet”, “Managing the Future Internet”, and “Business models, value chains and new players”. Six main drivers and challenges are presented in a scenario of evolution “Elephant and Gazelle”, from the business environment viewpoint. Then, four different business use cases are addressed: Network Virtualisation, New Ways of Information Delivery, Internet of Things and Community-Oriented Applications.
Jukka Salo, Luis M. Correia

Chapter 4. Network Design

A proposal is presented for a possible Architecture Framework covering certain demands of the Future Internet. Some concepts, terms and the basic constructs are defined, in order to model Network Architectures. This Architecture Framework provides two levels of views on network architectures: the macroscopic view, mainly focusing on structuring the network at a higher level of abstraction, and introducing the concept of Strata; the microscopic view, concentrating more on the functions needed in the network nodes, their selection and composition to Netlets. Functional Blocks are presented as the common points between the two views of the architecture. The Component Based Architecture constructs and principles are used as the basis to provide reusable frameworks that minimise the design and development times of new network architectures.
Susana Perez Sanchez, Roland Bless

Chapter 5. Naming and Addressing

Some Remarks on a Basic Networking Ingredient
A discussion of “names”, “addresses”, and their relation to each other from first principles is provided. We put this discussion into the context of the structure of communication systems, where we compare layered schemes vs. functionally complete ones. We identify the need for namespaces with proper information hiding and how to relate such namespaces to each other by a correctly interpreted name resolution concept. It turns out that this approach simplifies a number of aspects of communication system design; for example, it turns out that name resolution and neighbor discovery are essentially the same thing. A basic view is given on names, addresses, and compartments. Name resolution is presented as the centerpiece of the problem.
Holger Karl, Thorsten Biermann, Hagen Woesner

Chapter 6. Security Aspects and Principles

Rethinking the fundamental network architecture seems to be able to solve some known architectural security problems of the existing internet, but proposals are also investigated more thoroughly from the security angle overall. The information-centric approach of 4WARD is built on the concept of securing information rather than locations and paths used for information transit. Doing so, the security principles based on ownership and controlling access at the originating source become challenged. At the same time, moving intelligence into the network itself challenges the underlying assumption of having an Internet consisting of neutral, dumb, and fundamentally cooperating and trusting autonomous domains. 4WARD states the security principles necessary for dynamical management of virtualized, largely self-configuring entities having specific properties. The specific security implementation choices necessary for network design, transport, routing, lookup, privacy, accountability, caching and monitoring are part of the design process, for which 4WARD contributes functional descriptions and the concept of a design repository. 4WARD acknowledges and considers the business and governmental control interests that will heavily influence the security direction into which the future network evolves.
Göran Schultz

Chapter 7. Interdomain Concepts and Quality of Service

How We Interconnect Networks and How We Manage Quality of Service (QoS)
One of the key challenges for the Future Internet is the correct definition and implementation of the domain concept. The domain concept is introduced. The interconnection model of the Internet and of current mobile operators is analysed, addressing service ubiquity and interdomain concepts developed in the scope of 4WARD as well. Then, a new interconnection model is introduced, with requirements, principles, and peering models; architecture elements are addressed, together with interconnection in virtual networks. Special attention is devoted to the still to solve problem of Multidomain Quality of Service, namely The Inter-Provider QoS problem, new challenges and tools for QoS in the Future Internet, and QoS in a network virtualisation environment.
Pedro Aranda Gutiérrez, Jorge Carapinha

Chapter 8. Managing Networks

We propose a solution for management, In-Network Management, which is based on decentralization, self-organization, and autonomy of management processes. Its key idea is that management stations outside the network delegate management tasks to the network itself, supporting large-scale networks that self-configure, dynamically adapt to external events and allow for low-cost operation. We discuss challenges, benefits, and approaches to In-Network Management. We present an architectural framework suitable for different levels of embedding within the network elements. Examples of novel algorithms supporting real-time monitoring in a distributed manner are presented, and self-adaptation schemes for resource control are discussed; we outline real-time monitoring of network-wide metrics, group size estimation, data search, and anomaly detection. We conclude that robust, distributed algorithms can be devised for a multitude of management tasks without introducing excessive amounts of overhead in the networked devices.
Daniel Gillblad, Alberto Gonzalez Prieto

Chapter 9. How Connectivity Is Established and Managed

An architecture for data transmission that puts technological and administrative domains (compartments) in the role of the keeper of this shared information is described. Paths are established between communicating entities, basic functional blocks that reappear in different layers of the Internet. We explain how certain functions like routing, access control, and resource management are recurring in entities at all layers, and therefore allow an object oriented definition of entities and paths. Compartments and generic paths limit the scope within which state information needs to be kept consistent. Compartment layering is fundamentally different from the established ISO/OSI model and the chapter discusses several examples for the use of cooperation between more than the traditional two end points of a transmission.
Hagen Woesner, Thorsten Biermann

Chapter 10. How to Manage and Search/Retrieve Information Objects

We present the overall vision for a network of information, illustrating the fundamental ideas, and explaining the mechanisms currently under development that will bring about a major paradigm change in networking. After briefly reviewing relevant scenarios where the current host-centric approach to information storage and retrieval is ill-suited for, we introduce how a new networking paradigm emerges, by adopting the information-centric network architecture approach. We illustrate how information retrieval may look like in the future, emphasizing on the user perspective. We then put forward the architectural requirements for a network of information. A description of the mechanisms, the “nuts and bolts” so to speak, of the technologies that implement a network of information is provided. We describe a network of information operation, providing examples and highlighting performance improvement. A long-term view is taken, with a discussion on evolution.
Septimiu Nechifor

Chapter 11. Use Case—From Business Scenario to Network Architecture

One describes how 4WARD processes, concepts and technologies can be used and applied as to provide a suitable network architecture to support a futuristic business scenario. One describes the actors and the network environment of a futuristic business scenario, the ‘AdHoc Community’, which is followed by an analysis and extraction of a set of non-technical business-related requirements. Those are further analyzed and then mapped onto a set of technical requirements. The Design Process is then followed in order to derive a suitable network architecture made up of components and interfaces, which can be deployed into a physical network infrastructure. At the end, a discussion on design options is provided, as well as a comparison with a solution based on existing technologies.
Martin Johnsson, Anna Maria Biraghi

Chapter 12. Prototype Implementations

An overview of the developed prototypes from which some are also publicly available from the 4WARD project web site is given. The prototype implementations of the architecture framework concepts are presented, showing the application and interactions of those concepts to other architectures described in the book: the virtualization architecture, and the In-Network-Management architecture. Since network virtualization also needs verification at a larger scale, parts of the concepts have been developed for a large virtualization testbed, which is also presented. The Generic Path concept for routing, forwarding and transport, and the Network of Information prototypes are described. In addition, their combination is outlined which shows that they nicely complement each other. Last, but not least, an integrated prototype showing the combination of Generic Path and In-Network Management concepts is shown, giving special focus on QoS aspects.
Denis Martin, Martina Zitterbart

Chapter 13. Conclusions

The main conclusions from each chapter are presented, together with a critical perspective. At the end a view on migration paths is given. The deployment of the project results is put into four perspectives: extension of current networks by means of technology, e.g., it is possible to use results from self-management to enhance management capabilities; adding functionality by providing overlay/underlay and control, e.g., the Network of information can be applied on top of the current Internet; network virtualisation is a means of not only sharing network resources, but it can also be used as a migration path to new network architectures e.g. for specialised custom-tailored networks; deployment of a completely new network based on 4WARD’s architecture framework, but this has limited commercial applicability and can only be used for very specialised networks, e.g., sensor networks.
Henrik Abramowicz, Klaus Wünstel


Weitere Informationen