![x Preface
Disclaimer
This book has been produced in the context of the 4WARD Project. The research
leading to these results has received funding from the European Community’s Sev-
enth Framework Programme ([FP7/2007–2013]) under grant agreement n° 216041.
All information in this document is provided “as is” and no guarantee or warranty
is given that the information is fit for any particular purpose. The user thereof uses
the information at its sole risk and liability. For the avoidance of all doubts, the
European Commission has no liability in respect of this book, which is merely rep-
resenting the authors view.](https://image.slidesharecdn.com/26902744-250515032549-86731597/85/Architecture-And-Design-For-The-Future-Internet-4ward-Project-Signals-And-Communication-Technology-2011th-Edition-Luis-M-Correia-13-320.jpg)
![List of Figures
Fig. 2.1 The 4WARD System Model . . . . . . . . . . . . . . . . . . . . 17
Fig. 2.2 High-level view of 4WARD Architecture Framework . . . . . . 19
Fig. 2.3 High-level view of the 4WARD Design Process . . . . . . . . . 21
Fig. 2.4 INM relationship with Governance/Knowledge . . . . . . . . . . 22
Fig. 2.5 Virtualization ecosystem . . . . . . . . . . . . . . . . . . . . . . 23
Fig. 2.6 NetInf high-level architecture . . . . . . . . . . . . . . . . . . . 26
Fig. 3.1 Future Internet ecosystem . . . . . . . . . . . . . . . . . . . . . 30
Fig. 3.2 The international regulator landscape (source [7]) . . . . . . . . 33
Fig. 3.3 Extreme scenarios characterised by six drivers with uncertain
development . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
Fig. 3.4 Evolvement of markets in the regulated environment . . . . . . . 39
Fig. 3.5 Network virtualisation ecosystem . . . . . . . . . . . . . . . . . 41
Fig. 3.6 Rivalry in the value chain of virtualisation . . . . . . . . . . . . 43
Fig. 3.7 Different interconnection types and an interconnection case
(example) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Fig. 4.1 Sketch of the 4WARD Architecture Framework . . . . . . . . . 62
Fig. 4.2 Relationship between Netlets and Strata . . . . . . . . . . . . . 64
Fig. 4.3 A stratum, its internal structure and interfaces . . . . . . . . . . 65
Fig. 4.4 Outline of the End-Node Architecture . . . . . . . . . . . . . . . 67
Fig. 4.5 Design Process . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Fig. 4.6 Overview of a virtual network topology and substrate networks . 75
Fig. 4.7 Relationship between roles and resources . . . . . . . . . . . . . 75
Fig. 4.8 The VNet lifecycle: Process overview . . . . . . . . . . . . . . . 77
Fig. 4.9 The VNet lifecycle: Overview of interfaces . . . . . . . . . . . . 79
Fig. 4.10 A substrate node hosting different virtual nodes . . . . . . . . . 82
Fig. 5.1 The entity E1 in compartment C1 is bound to the name N2.1 of
E2 inside C2, making N2.1@C2 an address of E1 . . . . . . . . 97
Fig. 5.2 The entities E1 and E2 share a single node compartment . . . . . 98
Fig. 5.3 Address serve to find a way to a neighbor . . . . . . . . . . . . . 100
Fig. 5.4 Arrangement of entities inside a compartment for routing table
example from the perspective of entity E . . . . . . . . . . . . . 104
xxv](https://image.slidesharecdn.com/26902744-250515032549-86731597/85/Architecture-And-Design-For-The-Future-Internet-4ward-Project-Signals-And-Communication-Technology-2011th-Edition-Luis-M-Correia-27-320.jpg)
![16 M. Johnsson
1. Let 1000 Networks Bloom
We will explore a new approach to a multitude of networks: the best network for
each task, each device, each customer, and each technology. Unlike the multi-
tude we had in the past, where different incompatible technologies were compet-
ing with each other, we want to create a framework that will allow these many
networks to bloom as a family of interoperable networks coexisting and comple-
menting each other.
2. Let Networks Manage Themselves
The main limits of current technologies are the scaling up to very large network
sizes, and the needed human intervention which is associated with considerable
cost, errors and with an inherent slowness in reacting to changing network condi-
tions. What we would like to have is a management entity as an inseparable part
of the network itself, generating extra value in terms of guaranteed performance
in a cost effective way, and capable of adjusting itself to different network sizes,
configuration, and external conditions.
3. Let a Network Path Be an Active Unit
We want to consider a path as an active part of the network that controls itself
and provides customized transport services. An active path can provide resilience
and fail-over, offer mobility, simultaneously use multiple different sequences of
links, secure and compress transmitted data, and optimize its performance all by
itself.
4. Let Networks Be Information-Centric
Users are primarily interested in using services and accessing information, not
in accessing nodes that host information or provide services. Consequently, we
want to build a network as a network of information and services that may be
mobile and distributed. In such a network, the users just accesses items of interest
by their name while the data locations can be completely hidden.
These tenets, together with the understanding of the current situation of today’s
Internet, formed the main drivers for the definition of the 4WARD Technical Re-
quirements [1], which laid the foundation for technical work within the 4WARD
project. This work ultimately resulted in the 4WARD System Model, which is de-
scribed in the following section.
2.2 The 4WARD System Model
Figure 2.1 depicts the 4WARD System Model, which has been developed with the
Tenets and the 4WARD Technical Requirements [1] as main principal input. The
system model gives the necessary definitions, specifications, principles and guide-
lines for designing, building, deploying, and manage interoperable network archi-
tectures. For that purpose, the 4WARD System Model consists of an Architecture
Framework and a set of Architecture Pillars which provides the essential technolo-
gies in many of the network architectures anticipated and required for the future
networks, though it is possible to also deploy and use them in migration scenarios.
With the 4WARD System Model we expect significant efficiency gains in the de-](https://image.slidesharecdn.com/26902744-250515032549-86731597/85/Architecture-And-Design-For-The-Future-Internet-4ward-Project-Signals-And-Communication-Technology-2011th-Edition-Luis-M-Correia-45-320.jpg)
![20 M. Johnsson
configuration of horizontal strata to meet the performance criteria for a network.
The Governance Stratum also establishes and maintains relations and agreements
with other networks.
The Repository contains the set of Building Blocks and Design Patterns for the
composition of functionalities (i.e., to construct the strata and the netlets) for specific
network architectures, including best practices and constraints to ensure interoper-
ability between network architectures.
2.3.2 The Design Process
Evolution of today’s networks including the Internet suffers from the inability to be
extended in a consistent and reliable way while maintaining certain assured prop-
erties, such as security, quality of service, reliability even in the broader context.
Much effort has to be spent for standardization, development and regression testing
when introducing even minor feature improvements, before deploying them on a
network-wide basis. Upgrading of a large installed base of network elements means
a big technological challenge and financial risk to the network operator and service
provider.
4WARD has succeeded in setting up a design process that in the future will enable
new network designs to be developed, tested and deployed without impacting the
installed network basis, when based on this 4WARD architecture framework and
building upon the recent progress in network virtualization. The innovative 4WARD
network design process leverages advantages of model-driven software engineering
techniques and the experiences in design and composition of web services, based
on OSGI principles [2].
As shown in Fig. 2.3, the following phases are considered in the design process:
1. Requirements Analysis: Starting from the business idea and requirements, the
goal of this step is to decompose them into the high level functionalities that
should be realized by the architecture to be designed. The output of this phase is
mainly the identification of the macroscopic architectural view of Strata, a first
draft of the main network components, and the specification of technical require-
ments for further refinement of the architecture.
2. Abstract Service Design: During this phase, the technical requirements and the
high level functionalities derived from these will be turned into abstract func-
tionalities and ways how they can be composed, following generic principles
and design patterns. The result of this design phase is the specification of the
Netlets operating at node level, and the Strata that constitute the distribution of
functionalities across the network nodes.
3. The Component Design Phase focuses on the detailed specification and com-
position of the Functional Blocks (FBs) used to implement the specific function-
ality. This includes the specification of the interfaces, properties, and require-
ments/prerequisites of the FBs. The output of this phase is the detailed design of](https://image.slidesharecdn.com/26902744-250515032549-86731597/85/Architecture-And-Design-For-The-Future-Internet-4ward-Project-Signals-And-Communication-Technology-2011th-Edition-Luis-M-Correia-49-320.jpg)
![Mutta ei vain sitä teidän hirveätä valkoista Burgundiviiniänne,
huomautti Guido Tronka. Vielä tänäänkin ovat suoneni viime
istunnostamme lähtien kireinä kuin viulunjänteet.
— Siis sektiä [saksalainen kuohuviini, sampanjaa muistuttava] ja
portteria, rehellisesti puoleksi kumpiakin, ehdotti Fink.
Se kelpaa! huudahti pikku Lanzau…
Samanlaista hornanjuomaa sekin, valitti Zernitz.
Kellarimestari, viinuri, edeskäypä hoi! huusivat herrat ja tekivät
tilauksensa.
Tällävälin Anton keksi epätoivoisen keinon. Hän pistäytyi
huoneesta ulkona, antoi tarjoilijalle taalerin ja pyysi häntä
lämmittämään pikku takahuoneen hehkuvan kuumaksi, vähääkään
välittämättä toisten herrojen vastalauseesta. Itse hän istuutui niin
kauas uunista kuin mahdollista ja näki suureksi ilokseen Finkin
painautuvan aivan kiinni sen rautalieriöön. Piankin täytyi helteen
käydä hänelle rasittavaksi; silloin heittäisi hän takin päältään, kuten
hänen tapansa oli sellaisissa tilanteissa, ja sitten kävisi Antonille ehkä
mahdolliseksi ottaa punakantinen kirja takintaskusta Finkin itsensä
nähden.
Minullapa on teille suuri uutinen kerrottavana, alotti Tönnchen.
Oletteko nähnyt Tronkan Alicea, Fink?
Enkä, sanoi Fink, kaataen lasiinsa; onko se hevonen vai joku
naikkonen?
Hevonen tietystikin! huusi Tönnchen. Pah, jättäkää tallitakki
tänään kotiin, virkkoi Fink. Mutta tämä on vietävän totta! huudahti](https://image.slidesharecdn.com/26902744-250515032549-86731597/85/Architecture-And-Design-For-The-Future-Internet-4ward-Project-Signals-And-Communication-Technology-2011th-Edition-Luis-M-Correia-63-320.jpg)
Architecture And Design For The Future Internet 4ward Project Signals And Communication Technology 2011th Edition Luis M Correia
- 1. Architecture And Design For The Future Internet 4ward Project Signals And Communication Technology 2011th Edition Luis M Correia download https://ebookbell.com/product/architecture-and-design-for-the- future-internet-4ward-project-signals-and-communication- technology-2011th-edition-luis-m-correia-53805488 Explore and download more ebooks at ebookbell.com
- 2. Here are some recommended products that we believe you will be interested in. You can click the link to download. Practical Api Architecture And Development With Azure And Aws Design And Implementation Of Apis For The Cloud 1st Edition Thurupathan Vijayakumar https://ebookbell.com/product/practical-api-architecture-and- development-with-azure-and-aws-design-and-implementation-of-apis-for- the-cloud-1st-edition-thurupathan-vijayakumar-52850356 Architecture And Embodiment The Implications Of The New Sciences And Humanities For Design Harry Francis Mallgrave https://ebookbell.com/product/architecture-and-embodiment-the- implications-of-the-new-sciences-and-humanities-for-design-harry- francis-mallgrave-23300452 Center 16 Latitudes Architecture In The Americas Barbara Hoidn https://ebookbell.com/product/center-16-latitudes-architecture-in-the- americas-barbara-hoidn-50365242 The Routledge Companion For Architecture Design And Practice Established And Emerging Trends Mitra Kanaani https://ebookbell.com/product/the-routledge-companion-for- architecture-design-and-practice-established-and-emerging-trends- mitra-kanaani-5403334
- 3. The Routledge Companion To Games In Architecture And Urban Planning Tools For Design Teaching And Research Marta Brkovi Dodig https://ebookbell.com/product/the-routledge-companion-to-games-in- architecture-and-urban-planning-tools-for-design-teaching-and- research-marta-brkovi-dodig-84708922 The Value Of Urban Design A Research Project Commissioned By Cabe And Detr Tonbridge England 2001 Great Britain Commission For Architecture And The Built Environment https://ebookbell.com/product/the-value-of-urban-design-a-research- project-commissioned-by-cabe-and-detr-tonbridge-england-2001-great- britain-commission-for-architecture-and-the-built-environment-77187168 The Routledge Companion To Ecological Design Thinking Healthful Ecotopian Visions For Architecture And Urbanism Mitra Kanaani https://ebookbell.com/product/the-routledge-companion-to-ecological- design-thinking-healthful-ecotopian-visions-for-architecture-and- urbanism-mitra-kanaani-44736276 Effective Software Development For The Enterprise Beyond Domain Driven Design Software Architecture And Extreme Programming 1st Edition Tengiz Tutisani https://ebookbell.com/product/effective-software-development-for-the- enterprise-beyond-domain-driven-design-software-architecture-and- extreme-programming-1st-edition-tengiz-tutisani-50271520 The Art Of Hardware Architecture Design Methods And Techniques For Digital Circuits 2012th Edition Mohit Arora https://ebookbell.com/product/the-art-of-hardware-architecture-design- methods-and-techniques-for-digital-circuits-2012th-edition-mohit- arora-2510534
- 6. SIGNALS AND COMMUNICATION TECHNOLOGY For other titles published in this series, go to www.springer.com/series/4748
- 7. Luis M. Correia Henrik Abramowicz Martin Johnsson Klaus Wünstel Editors Architecture and Design for the Future Internet 4WARD Project Foreword by Joao Schwarz da Silva
- 8. Editors Prof. Luis M. Correia Technical University of Lisbon Instituto Telecomunicacoes Instituto Superior Tecnico Av. Rovisco Pais 1049-001 Lisbon, Portugal luis.correia@lx.it.pt Dr. Henrik Abramowicz Ericsson Research Isafjordsgatan 14E 16480 Stockholm Sweden Dr. Martin Johnsson Ericsson AB Torshamnsgatan 23 16480 Stockholm Sweden Dr. Klaus Wünstel Alcatel Lucent Bell Labs Lorenzstrasse 10 70435 Stuttgart Germany ISSN 1860-4862 ISBN 978-90-481-9345-5 e-ISBN 978-90-481-9346-2 DOI 10.1007/978-90-481-9346-2 Springer Dordrecht Heidelberg London New York © Springer Science+Business Media B.V. 2011 No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without written permission from the Publisher, with the exception of any material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work. Cover design: VTEX, Vilnius Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com)
- 9. Foreword The development of computing and resource sharing as we have known them un- til recently is about to radically change course as its center of gravity is shifting with technologies, service architectures allowing for applications to migrate to the cloud. This shift from Web 2.0 to Web 3.0 will give rise to an Internet of services of unprecedented scope and scale. We are now entering a new phase of ICT driven innovation and growth based on the Internet of Services which more and more will be accessible through what could be called the Mobile and Wireless Web. Already today applications of wireless technology are a major driver of economic value in the EU economy. These are estimated at 250 bn€ or 2–3% of GDP and rising. In the coming five years it is expected that close to 7 billion users or the entire planet’s population will have use of a mobile handset of which a great majority will be de- vices classified as smart-phones. This is an unprecedented development exceeding the diffusion rates of technolo- gies such as television or even pen and paper not only in terms of penetration and use but in its speed of take-up. We should expect an explosion of new applications with the potential to radically change the way in which we live and work. Examples are easy to cite: industrial and commercial applications in the supply chain, nomadic services for mobile workers, remote environmental monitoring or disaster and se- curity systems that save lives by putting essential information into the hands of first responders, health and education services. In such a remodelled world, new alliances will be created, new stakeholders will emerge, new modes of interaction will filter through into business practices, and new business models will proliferate. The Internet itself will no longer be a network of networks simply connecting computers and servers to become an Internet that connects “things” together: communicating devices by the billions, cars, machines of all sorts, household appliances, energy meters, windows, lights, etc. Around this new Internet will be borne a new economy of web based services and applications. There are two key implications of this new Internet. First, this new world wide web of “things that think” will create a sensory network that will allow a leap for- ward in the human knowledge about the world we live in. It will lend itself to all sorts of new applications such as energy efficiency, health and welfare services, ef- ficient transport and so on. If we do this well, there will be a massive improvement v
- 10. vi Foreword in our quality of life and sustainability, not just because of the services, not just be- cause of the competitive advantage of being an earlier mover, but because European values of openness and democracy will define the form that the Internet takes. Second we must liberate the economic potential of the single European market that is still locked up in fragmented national markets. In particular we must now strengthen the real economy by stimulating solid and sustainable business growth in high value goods and services that respond to real market needs. From the cur- rent period of uncertainty and as inevitable structural changes emerge it is essential to look for the growth opportunities in tomorrow’s world. The industrial and re- search community gathered around the Future Internet Assembly, has certainly the talent and the capability to shape the future. All it takes is the ambition to overcome fragmented markets and the will to build on our strengths by creating open single markets for innovative goods and services and by going for innovation and change. In creating the conditions that will allow Europe to benefit from the emerging economic opportunities, we must make sure that the Future Internet remains open. The key economic characteristic of the current Internet has been that it has created an unprecedentedly open platform for innovation and development of new services. We must keep this characteristic of openness by ensuring that open standards and eventually open-source software are the core of our actions. While the financial health of many companies worldwide is still based on proprietary models and gate- keeper business models, the world ahead of us will call for models whose economic basis offers a greater degree of liberty to the consumer or the enterprise. As the Future Internet unfolds before us, the need will arise to move toward smarter and greener infrastructures. This is a big challenge, but also a great oppor- tunity, because it will amplify the reach of the Internet to novel usages and industrial sectors. Indeed time has now come to go one step beyond what has been achieved so far. We must closely couple our Future Internet technology research and develop- ment with applications of high societal value such as health, urban mobility, energy grids or smart cities. In doing so, we will be able to provide an early “Internet re- sponse” to the many societal challenges with which we are confronted today. Multiple regional initiatives are currently emerging in view of defining the future Internet. Japan and Korea have made public their ambitious u-Japan and u-Korea initiatives, China is supporting the domain through an ambitious and integrated in- dustrial policy, in the US the GENI programme and facility is a key contributor to the debate on the future of the Internet. These initiatives are not all tackling the issue of the Internet evolution as part of their core objectives, but are certainly related to technological and socio-economic scenarios (ubiquity, connected devices) that will clearly need to be taken into account when addressing the Internet of Tomorrow. From an EU perspective, it would be beneficial to build on these various ini- tiatives and create the conditions that would bring about a closer complementarity and cooperation between all actors associated to the definition, testing and valida- tion work. One of the main objectives of multilateral partnerships should be the emergence of global standards. Standards are indeed a key element to achieve in- teroperability and openness, two of the essential Internet characteristics that have contributed to its success. Indeed the ever growing multiplicity of players as well as
- 11. Foreword vii the convergence of different sectors has lead to increased complexities in the stan- dards making processes as illustrated by debates on IPR portfolios, as well as on the degree of openness, transparency and access. Early co-operation and international partnerships on novel technologies are hence key to facilitate broader consensus, early agreements on standards by the key players while holding the promise to alleviate subsequent IPR disputes. An important point to note is that the new economy created by the Internet is producing beyond a business revolution a unique opportunity to generate enormous environmental benefits, particularly if the right technological choices are made at the level of the infrastructure. In addition by reducing the amount of energy and materials consumed by business and by increasing overall productivity, the new Internet holds the promise to revolutionize the relation between economic growth and the environment. It is in the above context that I have the pleasure to share with you my satisfaction as to the achievements of the EU RD Project 4WARD. The book you are about to read, details the many unique contributions of the project to the development of a solid scientific basis for the Future Internet. Key amongst its many contributions are those relating to a new architecture framework where mobility, multi-homing and security become an intrinsic part of the network architecture rather than add-on solu- tions, hence allowing networks to bloom as a family of interoperable networks each complementing each other and each addressing individual requirements such as mo- bility, QoS, security, resilience, wireless transport and energy-awareness. 4WARD also addressed particularly well the question as to how virtualization can provide an opportunity to roll out new architectures, protocols, and services with network service providers sharing a common physical infrastructure. Tightly coupled to vir- tualization is network management, where 4WARD has broken new territory by advocating an approach where management functions come as embedded capabili- ties of devices. 4WARD has gone further than others by recognizing the paradigm shift brought about by the move from a node-centric age to an information-centric age. The partners and scientific staff of 4WARD are to be congratulated for the work performed and for providing a perfect illustration of how Europe’s commitment and creativity will enable the future. Dr. Joao Schwarz da Silva Former Director of DG-INFSO, European Commission
- 12. Preface The current Internet is a tremendous commercial success and has become widely spread after having started as an academic research network to become a network for the everyday life for ordinary people. The Internet of today has its origins from the 70-ties, and was essentially simple but open for new applications and designed for the fixed network. It is however been increasingly challenged by the new trans- mission technologies based on radio and fiber, as well as by the new applications and media types that increasingly rely on overlays to make up for shortages in the core Internet architecture. In particularly, the even greater success of mobile networks has questioned the current Internet, which has reached a state of high complexity with regard to support of mobility, interoperability, configuration and management and vulnerability in an untrustworthy world. The project 4WARD, started January 2008 and completed by June 2010, had the task to research on Architecture and Design for a Future Internet. The project took a clean slate research approach, which means that in its research it was not constrained by the current Internet. It does not mean however that the project favored a clean slate deployment, but rather saw a migration approach in how to apply its research results into the current Internet. The project was partly EU funded under the EU Framework Programme 7 and consists of the 33 partners (see Appendix). There have been over 120 persons in the project, and for this reason it is not possible to list all that have contributed to the project and results. We would however like to acknowledge all for their valuable contributions. Further to that, we would like to acknowledge the help and support the project has experienced by the project officer Dr. Paulo de Sousa and the good collaboration we have had. The work of Daniel Sebastiao (IST, Lisbon) in the edit- ing work is also acknowledged. This book describes the salient results out of this project and covers not only technical results but deals also with socio-economic issues. The Editors ix
- 13. x Preface Disclaimer This book has been produced in the context of the 4WARD Project. The research leading to these results has received funding from the European Community’s Sev- enth Framework Programme ([FP7/2007–2013]) under grant agreement n° 216041. All information in this document is provided “as is” and no guarantee or warranty is given that the information is fit for any particular purpose. The user thereof uses the information at its sole risk and liability. For the avoidance of all doubts, the European Commission has no liability in respect of this book, which is merely rep- resenting the authors view.
- 14. Contents 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Luis M. Correia, Henrik Abramowicz, Martin Johnsson, and Klaus Wünstel 2 A System Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Martin Johnsson 3 Socio-economic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Jukka Salo and Luis M. Correia 4 Network Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 Susana Perez Sanchez and Roland Bless 5 Naming and Addressing . . . . . . . . . . . . . . . . . . . . . . . . . 89 Holger Karl, Thorsten Biermann, and Hagen Woesner 6 Security Aspects and Principles . . . . . . . . . . . . . . . . . . . . . 115 Göran Schultz 7 Interdomain Concepts and Quality of Service . . . . . . . . . . . . . 133 Pedro Aranda Gutiérrez and Jorge Carapinha 8 Managing Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 Daniel Gillblad and Alberto Gonzalez Prieto 9 How Connectivity Is Established and Managed . . . . . . . . . . . . 173 Hagen Woesner and Thorsten Biermann 10 How to Manage and Search/Retrieve Information Objects . . . . . . 201 Septimiu Nechifor 11 Use Case—From Business Scenario to Network Architecture . . . . 225 Martin Johnsson and Anna Maria Biraghi xi
- 15. xii Contents 12 Prototype Implementations . . . . . . . . . . . . . . . . . . . . . . . 245 Denis Martin and Martina Zitterbart 13 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279 Henrik Abramowicz and Klaus Wünstel Appendix Project Description and Reports . . . . . . . . . . . . . . . . 293 Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305
- 16. Contributors Editors Henrik Abramowicz Ericsson Research, Stockholm, Sweden Pedro Aranda Gutiérrez Telefonica I+D, Madrid, Spain Thorsten Biermann University of Paderborn, Paderborn, Germany Anna Maria Biraghi Telecom Italia, Turin, Italy Roland Bless Karlsruhe Institute of Technology, Karlsruhe, Germany Jorge Carapinha PT Inovação, Aveiro, Portugal Luis M. Correia IST/IT—Technical University of Lisbon, Lisbon, Portugal Daniel Gillblad SICS—Swedish Institute of Computer Science, Stockholm, Swe- den Alberto Gonzalez Prieto KTH—Royal Institute of Technology, Stockholm, Swe- den Martin Johnsson Ericsson Research, Stockholm, Sweden Holger Karl University of Paderborn, Paderborn, Germany Denis Martin Karlsruhe Institute of Technology, Karlsruhe, Germany Septimiu Nechifor Siemens, Brasov, Romania Susana Perez Sanchez Tecnalia-Robotiker, Zamudio (Vizcaya), Spain Jukka Salo Nokia Siemens Networks, Espoo, Finland Göran Schultz Ericsson Research, Jorvas, Finland Hagen Woesner Technical University of Berlin EICT, Berlin, Germany Klaus Wünstel Alcatel Lucent Bell Labs, Stuttgart, Germany Martina Zitterbart Karlsruhe Institute of Technology, Karlsruhe, Germany xiii
- 17. xiv Contributors Other Contributors Alexander Landau, Anders Eriksson, Andrei Bogdan Rus, Anghel Botos, Asan- ga Udugama, Avi Miron, Bengt Ahlgren, Björn Grönvall, Bogdan Tarnauca, Börje Ohlman, Carmelita Gorg, Chris Foley, Christian Dannewitz, Christian Tschu- din, Christoph Werle, Daniel Horne, Daniel Sebastião, Djamal Zeghlache, Do- minique Dudowski, Eric Renault, Fabian Wolff, Fabrice Guillemin, Fetahi Wuhib, Gabriel Lazar, Georgeta Boanea, Gerhard Hasslinger, Giorgio Nunzi, Gorka Her- nando Garcia, Ian Marsh, Jim Roberts, João Gonçalves, Jovan Golić, Jukka Mäkelä, Karl Palmskog, Kostas Pentikousis, Lars Voelker, Laurent Mathy, Leonard Pitzu, Liang Zhao, M. Ángeles Callejo Rodríguez, Mads Dam, Marco Marchisio, Mar- cus Brunner, Mario Kind, Matteo D’Ambrosio, Melinda Barabas, Michael Kleis, Miguel Ponce de Leon, Mohammed Achemlal, Olli Mämmelä, Ove Strand- berg, Panagiotis Papadimitriou, Patrick Phelan, Rebecca Steinert, René Rembarz, Reuven Cohen, Rolf Stadler, Rui Aguiar, Sabine Randriamasy, Teemu Rautio, Thomas Monath, Thomas-Rolf Banniza, Vinicio Vercellone, Virgil Dobrota, Yasir Zaki, Zakaria Khan, Zsolt Polgar, Zsuzsanna Kiss
- 18. List of Acronyms 3G Third Generation 3GPP 3rd Generation Partnership Project 4G Fourth Generation AAA Administration, Authorization, and Authentication ACK Acknowledgment AdHC Ad-Hoc Communities AHDR Ad-Hoc Disaster Recovery AM Anchorless Mobility AN Access Node AODV Ad-Hoc On-Demand Vector AP Access Point API Application Programming Interface ARP Address Resolution Protocol ARQ Automatic Repeat Request AS Autonomous System ASN Autonomous System Number BE Best Effort BER Bit Error Rate BEREC Body of European Regulators BFD Bidirectional Forwarding Detection BGP Border Gateway Protocol BIOS Basic Input/Output System BLER Block Error Rate BO Bit-level Objects BU Binding Update CA Channel Assignment CAIDA Cooperative Association for Internet Data Analysis CAPEX Capital Expenditure CBA Component Based Architecture CBR Constant Bit Rate CBSE Component Based Software Engineering xv
- 19. xvi List of Acronyms CCFW Cooperation and Coding Framework CCN Content Centric Networks CEP Connected End Points CF Cooperation/Coding Facility CFL CF Layer CLQ Cross-Layer QoS CMT Concurrent Multipath Transfer CN Correspondent Node Co-AD Content Adaptation CPU Central Processing Unit CRC Cyclic Redundancy Check CSMA Carrier Sense Multiple Access CSMA/CD Carrier Sense Multiple Access/Collision Detection CT Compartment CTR Compartment Record DA Deviation Advertisement DBA Dynamic Bandwidth Allocation DCF Dispersion Compensating Fiber DDOS Distributed Denial of Service DF Digital Fountain DGE Dynamic Gain Equalizers DHCP Dynamic Host Configuration Protocol DHT Distributed Hash Table DIF Distributed IP Facility DL Downlink DMA Dynamic Mobility Anchoring DMV2 Data-Multimedia-Voice-Video DNC Deterministic Network Coding DNS Domain Name System DONA Data Oriented Network Architecture DOS Denial of Service DSL Domain Specific Language, or Digital Subscriber Line DTN Delay/Disruption Tolerant Network E2E End-to-End EC European Commission ECN Explicit Congestion Notification EFCP Error and Flow Control Protocol EGP Exterior Gateway Protocol EMT Emergency Medical Team EP End Point EPON Ethernet Passive Optical Network ERC Emergency Response Command ERG European Regulators Group ETT Expected Transmission Time ETX Expected Transmission Count
- 20. List of Acronyms xvii FARA Forward Directive, Association, and Rendezvous Architecture FB Functional Block FCAPS Fault, Configuration, Accounting, Performance, and Security FDP Forwarding Decision Process FEC Forward Error Correction FER Frame Error Rate FI Future Internet FIA Future Internet Architectures FIB Forwarding Information Base FIFO First-In First-Out discipline FIM Flow Interception Module FIND The future Internet design FL Folding Link Fl-EP Flow Endpoint Fl-RO Flow Routing FN Folding Node FNE Forwarding NE FO Fixed Operator ForCES Forwarding/Control Element Separation FP7 Framework Programme 7 FPNE Flow Processing NE FQ Fair Queuing FRR Fast Reroute FSA Flow State Advertisement FTP File Transfer Protocol FTTH Fiber to the Home GAP Generic Aggregation Protocol GEF Graphical Editing Framework GENI Global Environment for Network Innovation GF Galois Field GGAP Gossip-Generic Aggregation Protocol GMOPR Grid MOPR GMP Global Management Point GMPLS Generalized Multi-Protocol Label Switching GMPR Generic Path Master Record GP Generic Path GPMR Generic Path Management Record GPRS General Packet Radio Service GPS Global Positioning System GRDF Generic Resource Description Framework GRX GPRS Roaming Exchange GRX GSM Roaming Exchange GSM Global System for Mobile Communications GSMA GSM Association GS-Node Governance Stratum Node
- 21. xviii List of Acronyms GUI Graphical User Interface HA Home Agent HEN Heterogeneous Experimental Network HIP Host Identity Protocol HTTP Hypertext Transfer Protocol iAWARE Interference Aware routing metric ICANN Internet Corporation for Assigned Names and Numbers ICN Information-Centric Network ICVNet Interconnecting Virtual Network ID Identifier IDR Inter Domain Routing IETF Internet Engineering Task Force IGP Interior Gateway Protocol ILA Interference-Load Aware routing metric ILC Inter Layer Communication ILR Inter Layer Routing ILS Information Lookup Service INI Information Network Interface INM In-Network Management InP Infrastructure Provider IO Information Object IOLS Information Object Lookup Service IP Internet Protocol IPC Inter-Process Communication IPTV Internet Protocol Television IPv6 Internet Protocol version 6 IPX IP packet eXchange ISP Internet Service Provider IT Information Technologies ITU International Telecommunication Union IXP Internet Exchange Point JSIM JavaSim KS-Node Knowledge Stratum Node LAN Local Area Network LLC Late Locator Construction LLID Logical Link ID LQO Link Quality Ordering LQODV Link Quality Ordering-based Distance Vector LSA Link State Advertisement LSP Label Switched Path LSR Label Switch Router LT Luby Transform LTE Long Term Evolution MAC Media Access Control MANET Mobile Ad Hoc Network
- 22. List of Acronyms xix MAP Mesh Access Point MBMS Multimedia Broadcast/Multicast Service MC Management Capabilities MDHT Multiple Distributed Hash Table MED Multi-Exit Discriminator MEE-GP Multihomed End-to-End GP MEEM Multihomed End-to-End Mobility MIC Metric of Interference and Channel-switching MIH Media Independent Handover MILP Mixed Integer Linear Program MIP Mobile IP MMS Multimedia Messaging System MN Mobile Node MNE Mediating NE Mo-AH Mobility Anchor MOPR Multi-Objective MPR MP Mediation Point MP2MP Multipoint-To-Multi-Point MP2P Multipoint-To-Point MP-BGP Multi-Protocol BGP MPC Multi-Party Computation MPLS Multiprotocol Label Switching MPR Multi-Path Routing MPR-CT MPR Compartment MPR-GP MPR GP MPR-ME MPR Master Entity MR Master Record MS-Node Machine Stratum Node MTU Maximum Transfer Unit NACK Negative Acknowledgment NAT Network Address Translation NATO! Not all at once NC Network Coding NDL Network Description Language NE Networking Entity NED NEtwork Description language (of OMNeT++) NetInf Network of Information NGN New Generation Network NHLFE Next Hop Label Forwarding Entry Ni-IO NetInf Information Object Ni-MG NetInfo Manager NIN NetInf node NLRI Network Layer Reachable Information Node CT Node Compartment NR Name Resolution
- 23. xx List of Acronyms NRS Name Resolution Service NSF National Science Foundation NSIS Next Steps in Signalling NSLP NSIS Signalling Layer Protocol NTP Network Time Protocol NW Network OADM Optical Add Drop Multiplexers OCS Optical Circuit Switching OD Origin Destination OFDM Orthogonal Frequency Division Multiplex OLA Optical Line Amplifiers OLT Optical Line Terminal OM Observation Module ONU Optical Network Unit OPEX Operational Expenditure OS Operating System OSGi Open Services Gateway initiative OSI-SM Open Systems Interconnection—System Management OSPF Open Shortest Path First OSPF-TE Open Shortest Path First—Traffic Engineering Extensions OSS Open-Source Software OTN Optical Transport Network OWL Web Ontology Language OXC Optical Cross-Connect P2MP Point-to-Multipoint P2P Peer to Peer Pa-EP Path Endpoint Pa-RO Path Routing PC Personal Computer PCE Path Computation Element (P)CN Congestion and Pre-Congestion Notification PDU Protocol Data Unit PER Packet Error Rate PHY Physical Layer PIM-DM Protocol Independent Multicast—Dense Mode PIM-SM Protocol Independent Multicast—Sparse Mode PMD Polarization Mode Dispersion PMS Personal Mobile Scenario PN Provisioning Network PnP Plug-and-Play PNP Physical Network Provider Po-EN Policy Engine PON Passive Optical Network PPP Point-to-Point PSR Packet Success Rate
- 24. List of Acronyms xxi PSTN Public Switched Telephone Network QoE Quality of Experience QoS Quality of Service RD Research and Development RA Resource Advertisement RAID Redundant Array of Inexpensive Disks RDF Resource Description Framework RDL Resource Description Language RDP Routing Decision Process RESCT Resolution Compartment RFC Request For Comments RFID Radiofrequency Identification RI Routing Instruction RIB Routing Information Base R-MAC Radio Medium Access Control RNC Random Network Coding RNE Routing NE RNG Random Number Generator RO Routing Object ROADM Reconfigurable Optical Add-Drop Multiplexer RSVP Resource Reservation Protocol RSVP-TE Resource Reservation Protocol—Traffic Engineering RTT Round-Trip Time RUI Routing Update Interval SA Service Agent SAP Service Access Point SATO Service-Aware Transport Overlay SCTP Stream Control Transmission Protocol SDH Synchronous Digital Hierarchy SE Self-managing Entities SGP Service Gateway Point, or Stratum Gateway Point SHIM6 Site Multihoming by IPv6 Intermediation SIM Subscriber Identity Module SINR Signal to Interference and Noise Ratio SIP Session Initiation Protocol SLA Service Level Agreement Sl-MA Service Level Agreement Manager SNMP Simple Network Management Protocol SNR Signal to Noise Ratio SOA Service Oriented Architecture SOCP Second Order Conic Program SON Service Oriented Networks SONET Synchronous Optical Network SP Service Provider SQF Shortest Queue First discipline
- 25. xxii List of Acronyms SRDF Semantic Resource Description Framework SRMF Semantic Resource Management Framework sRTT smoothed Round-Trip Time SSDP Simple Service Discovery Protocol SSP Service Stratum Point, or Stratum Service Point SVN SubVersioN SW Software Tagg-GP Transport aggregate GP TCG Trusted Computing Group TCP Transmission Control Protocol TDM Time Division Multiplex TDMA Time Division Multiple Access TE Traffic Engineering TENE Traffic Engineering NE TGP Transport GP TIC Time Interval Counter TLS Transport Layer Security TM Transformation Module TMN Telecommunications Management Network TO Time-out To-DB Topology Database Tr-MO Traffic Monitoring TTL Time to Live TTM Time To Market TV Television UA User Agent UDP User Datagram Protocol UIP Unmanaged Internet Protocol UL Uplink UML Unified Modelling Language UMTS Universal Mobile Telecommunications System UPnP Universal Plug and Play URL Uniform Resource Locator VBR Variable Bit Rate Vi-Node Virtual Node VLAN Virtual Local Area Network VLC VLC media player VNet Virtual network VNM VNet Management VNO VNet Operator VNP VNet Provider VoD Video on Demand VoIP Voice over IP VPN Virtual Private Network WAN Wide Area Network
- 26. List of Acronyms xxiii WAP Wireless Access Point WCETT Weighted Cumulative ETT WDM Wavelength Division Multiplexing WFQ Weighted Fair Queuing discipline WiFi Wireless Fidelity WiMAX Worldwide Interoperability for Microwave Access WLAN Wireless Local Area Network W-LLC Wireless Link Layer Control WMAN Wireless Metropolitan Area Networks WMN Wireless Mesh Network WMOPR Wireless MOPR WMVF Wireless Medium Virtualization Framework WP Work Package xDSL Digital Subscriber Line XML eXtensible Markup Language
- 27. List of Figures Fig. 2.1 The 4WARD System Model . . . . . . . . . . . . . . . . . . . . 17 Fig. 2.2 High-level view of 4WARD Architecture Framework . . . . . . 19 Fig. 2.3 High-level view of the 4WARD Design Process . . . . . . . . . 21 Fig. 2.4 INM relationship with Governance/Knowledge . . . . . . . . . . 22 Fig. 2.5 Virtualization ecosystem . . . . . . . . . . . . . . . . . . . . . . 23 Fig. 2.6 NetInf high-level architecture . . . . . . . . . . . . . . . . . . . 26 Fig. 3.1 Future Internet ecosystem . . . . . . . . . . . . . . . . . . . . . 30 Fig. 3.2 The international regulator landscape (source [7]) . . . . . . . . 33 Fig. 3.3 Extreme scenarios characterised by six drivers with uncertain development . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Fig. 3.4 Evolvement of markets in the regulated environment . . . . . . . 39 Fig. 3.5 Network virtualisation ecosystem . . . . . . . . . . . . . . . . . 41 Fig. 3.6 Rivalry in the value chain of virtualisation . . . . . . . . . . . . 43 Fig. 3.7 Different interconnection types and an interconnection case (example) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Fig. 4.1 Sketch of the 4WARD Architecture Framework . . . . . . . . . 62 Fig. 4.2 Relationship between Netlets and Strata . . . . . . . . . . . . . 64 Fig. 4.3 A stratum, its internal structure and interfaces . . . . . . . . . . 65 Fig. 4.4 Outline of the End-Node Architecture . . . . . . . . . . . . . . . 67 Fig. 4.5 Design Process . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 Fig. 4.6 Overview of a virtual network topology and substrate networks . 75 Fig. 4.7 Relationship between roles and resources . . . . . . . . . . . . . 75 Fig. 4.8 The VNet lifecycle: Process overview . . . . . . . . . . . . . . . 77 Fig. 4.9 The VNet lifecycle: Overview of interfaces . . . . . . . . . . . . 79 Fig. 4.10 A substrate node hosting different virtual nodes . . . . . . . . . 82 Fig. 5.1 The entity E1 in compartment C1 is bound to the name N2.1 of E2 inside C2, making N2.1@C2 an address of E1 . . . . . . . . 97 Fig. 5.2 The entities E1 and E2 share a single node compartment . . . . . 98 Fig. 5.3 Address serve to find a way to a neighbor . . . . . . . . . . . . . 100 Fig. 5.4 Arrangement of entities inside a compartment for routing table example from the perspective of entity E . . . . . . . . . . . . . 104 xxv
- 28. xxvi List of Figures Fig. 5.5 Scenario for the resolution table example . . . . . . . . . . . . . 105 Fig. 6.1 Example multidomain network . . . . . . . . . . . . . . . . . . 128 Fig. 7.1 Peering types through Internet Exchange Point (IXP) and using private peering . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 Fig. 7.2 Peering in the stratum model . . . . . . . . . . . . . . . . . . . 141 Fig. 7.3 Netlet types: components needed to provide interoperation at the Netlet level . . . . . . . . . . . . . . . . . . . . . . . . . . . 141 Fig. 7.4 The Folding Link concept . . . . . . . . . . . . . . . . . . . . . 142 Fig. 7.5 The Folding Node concept . . . . . . . . . . . . . . . . . . . . 142 Fig. 7.6 QoS in a multi-domain scenario . . . . . . . . . . . . . . . . . . 144 Fig. 7.7 Single virtual network domain, multiple infrastructure domains . 147 Fig. 8.1 Transitional diagram: three dimensions of the functional design space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156 Fig. 8.2 Main actors in INM . . . . . . . . . . . . . . . . . . . . . . . . 157 Fig. 8.3 Overview of the INM framework . . . . . . . . . . . . . . . . . 158 Fig. 8.4 Comparison of gossip-based vs. tree-based algorithms, in terms of robustness and performance. Figure shows estimation error vs. protocol overhead . . . . . . . . . . . . . . . . . . . . . . . 161 Fig. 8.5 Performance of detection and localization of communication failures and latency variations . . . . . . . . . . . . . . . . . . . 163 Fig. 8.6 Least congested path first scenario . . . . . . . . . . . . . . . . 169 Fig. 9.1 Overview and interaction of GP architecture components . . . . 175 Fig. 9.2 Mediation points are a union of several Entities. The data path, connecting Endpoints using a single ForMuxer, is not shown . . 178 Fig. 9.3 Shape of the service graphs of different architectural options . . 182 Fig. 9.4 Realization of an HTTP GP . . . . . . . . . . . . . . . . . . . . 185 Fig. 9.5 UML class diagram of Compartment and Generic Path class. Note the recursive composition of GP . . . . . . . . . . . . . . . 188 Fig. 9.6 Overview of the CCFW components in a classic layered networking system . . . . . . . . . . . . . . . . . . . . . . . . . 190 Fig. 9.7 Implementing traffic transformation with a special GP. The existing realization of the top GP is changed during its run time . 191 Fig. 9.8 Implementing traffic transformation with an additional GP. The existing realization of the top GP is changed during its run time . 192 Fig. 9.9 The Dynamic Mobility Anchoring (DMA) scheme . . . . . . . . 194 Fig. 9.10 The Anchorless Mobility (AM) scheme . . . . . . . . . . . . . . 195 Fig. 9.11 Multihomed End-to-End GP . . . . . . . . . . . . . . . . . . . . 196 Fig. 9.12 Handover steps . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 Fig. 10.1 Information centric networking . . . . . . . . . . . . . . . . . . 203 Fig. 10.2 NetInf architecture overview . . . . . . . . . . . . . . . . . . . 206 Fig. 10.3 A conceptual Information model . . . . . . . . . . . . . . . . . 209 Fig. 10.4 Information model IO format . . . . . . . . . . . . . . . . . . . 212 Fig. 10.5 NetInf network architecture . . . . . . . . . . . . . . . . . . . . 215 Fig. 10.6 A NetInf mobile multi-access scenario . . . . . . . . . . . . . . 219 Fig. 11.1 AdHoc Community scenario overview . . . . . . . . . . . . . . 228
- 29. List of Figures xxvii Fig. 11.2 Business model for AdHoc Community . . . . . . . . . . . . . . 230 Fig. 11.3 The specification of the horizontal and vertical strata and their respective functionalities and protocols . . . . . . . . . . . . . . 237 Fig. 11.4 Netlet design . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240 Fig. 12.1 Screenshot of Netlet Editor showing two video codec Netlets built with different building blocks . . . . . . . . . . . . . . . . 247 Fig. 12.2 VNet Management (VNM) . . . . . . . . . . . . . . . . . . . . 250 Fig. 12.3 Agent architecture and physical server configuration . . . . . . . 251 Fig. 12.4 Physical router and switch configuration . . . . . . . . . . . . . 252 Fig. 12.5 Network virtualization setup: Node Architecture Daemon running inside a virtual node, Virtualization demo testbed . . . . 254 Fig. 12.6 Prototype overview . . . . . . . . . . . . . . . . . . . . . . . . 256 Fig. 12.7 Resource discovery with limited information . . . . . . . . . . . 257 Fig. 12.8 Virtual link setup . . . . . . . . . . . . . . . . . . . . . . . . . 258 Fig. 12.9 Management access to virtual nodes . . . . . . . . . . . . . . . 259 Fig. 12.10 Control of objectives through the INM Framework . . . . . . . . 263 Fig. 12.11 Reference scenarios for application of the prototype in the future Internet . . . . . . . . . . . . . . . . . . . . . . . . . . . 264 Fig. 12.12 Inheritance graph for Entity and Core implementations. In this example, all Entities implementing protocols (leaves of the tree) are environment-specific as they require some functions of a certain execution environment (OMNeT++ LAN interface and POSIX TCP/UDP stack) . . . . . . . . . . . . . . . . . . . . . . 267 Fig. 12.13 A Network of Information consisting of NetInf nodes . . . . . . 268 Fig. 12.14 CLQ-based testbed used for Generic Path . . . . . . . . . . . . . 272 Fig. 12.15 Used transfer rate for the flow R1-R5-R6-R4 . . . . . . . . . . . 274
- 30. List of Tables Table 5.1 Routing table of an entity E in a compartment C1 . . . . . . . . 104 Table 5.2 Resolution table from the perspective of node compartment 2 of Fig. 5.5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 Table 5.3 Resolution configuration table . . . . . . . . . . . . . . . . . . 110 Table 9.1 Routing table of an entity E1 in a compartment C1 . . . . . . . . 183 Table 9.2 Name resolution table of an Entity E1 in a compartment C1 . . . 184 Table 11.1 Business requirements . . . . . . . . . . . . . . . . . . . . . . . 233 Table 11.2 Technical requirements . . . . . . . . . . . . . . . . . . . . . . 235 Table 11.3 Selection of components for the implementation of each Functional Block for the Ad-Hoc Communities example . . . . . 241 xxix
- 31. Chapter 1 Introduction Luis M. Correia, Henrik Abramowicz, Martin Johnsson, and Klaus Wünstel Abstract It starts by addressing some of the problems with current Internet net- works, its core architecture and its evolution model. Current architectures are typi- cally 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 busi- ness 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. 1.1 Problems with Current Internet The discussion on the “Network of the Future” is gaining in intensity due to in- creasing concerns about the inability of the current Internet to address a number of important issues affecting present and future services and to the impetus provided by “clean slate design” research initiatives launched in the US, Europe and Asia. L.M. Correia () IST/IT—Technical University of Lisbon, Lisbon, Portugal e-mail: luis.correia@lx.it.pt H. Abramowicz · M. Johnsson Ericsson Research, Stockholm, Sweden K. Wünstel Alcatel Lucent Bell Labs, Stuttgart, Germany L.M. Correia et al. (eds.), Architecture and Design for the Future Internet, Signals and Communication Technology, DOI 10.1007/978-90-481-9346-2_1, © Springer Science+Business Media B.V. 2011 1
- 32. 2 L.M. Correia et al. Many problems with the current network architecture have been recognised for a long time, but have not received a satisfactory solution. Below is a description of a number of issues pertaining to the current Internet. It is just example of problems, and not an exhaustive list of issues. It should be remembered that the current Internet was initially developed for a world in which a limited number of trusted nodes interconnected by copper based transmission technology implemented distributed applications mostly some kind of file transfer and message exchange. The initial architecture developed for this pur- pose was essentially simple, but open for new applications. Its evolution has led to a tremendous success—the Internet as we know it today. It is however far from clear that it is still the optimally evolvable solution, able to meet the challenges of dominating fibre optics and radio transmission technology, real-time multimedia and file-sharing applications and exposure to an untrustworthy world. Furthermore the Internet, starting as a simple set of protocols and rules, has over the decades reached a state of high complexity with regard to interoperability, configuration and management. 1.1.1 Internet’s Core Architecture and Evolution Model Is No Longer Suitable The predecessor of today’s Internet, the Arpanet, started in the late 1960s as a net- work of four university networks. It was created as a fixed network for hosts that were neither mobile nor wirelessly connected. Today, firewalls, network address and port translators, as well as session border controllers decouple the different IP networks at various layers. The capability of all end nodes to act as both consumer and producer of information has been continually reduced and, in the case of mo- bile nodes, may be considered not to exist at all; IPv6 failed to provide the necessary general architectural overhaul the Internet needs to become the universal Network of the Future that people can rely on. Worse, there are good reasons to believe that a single network solution will not be able to cover and satisfy the future needs in networking. While the scale of the Internet has not yet reached its limit, the growth in func- tionality, i.e., the ability of the global system to adapt to new functional require- ments, has almost come to a standstill. We have reached a critical point in an im- pressive development cycle that now requires a major change. 4WARD is taking a long term perspective on network (r)evolution and adopts a clean-slate research approach. We expect this research effort to have major industrial impact during the next decade. 1.1.2 Ossification of Internet Current architectures for communication systems are typically developed around layered models (e.g., Internet, OSI, 3GPP). Practice in the open Internet environ-
- 33. 1 Introduction 3 ment has shown that it is difficult to deploy network enhancements, such as IPv6, IPSec, MobileIP, or multicast. The dramatic growth of the Internet has also brought into sharp focus its architectural deficiencies such as lack of support for QoS and seamless mobility, security vulnerabilities, address shortage, to name a few. Al- though a number of solutions have been proposed for these problems, these can, at best, be described as a patchwork of fixes to fill architectural holes. Furthermore, most of them have come about in an uncoordinated and ad-hoc fashion and hence, they have spawned problems of their own. The resulting system has become quite complex, often with similar functionality re-appearing over and over again in different protocols and layers. The piecemeal ad hoc approach to solving problems that applies patches to certain parts of the protocol stack can in fact jeopardise the operation and performance of other parts of the communication system. 1.1.3 Surge of Unwanted Traffic, Including but Not Limited to SPAM The various forms of unwanted traffic, including spam, distributed denial of service (DDoS), and phishing, are arguably the biggest problem in the current Internet. Most of us receive our daily dosage of spam messages; the more lucky of us just a few of them, the more unlucky a few hundreds each day. Distributed denial of service attacks are an everyday problem to the large ISPs, with each major web site or content provider getting their share. And, as we all know, phishing is getting increasingly common and cunningly sophisticated. The root reasons to unwanted traffic seem to be best characterised with eco- nomics. We can characterise the current Internet as a global, distributed message passing system where the recipient pays the main cost of unwanted communica- tion. This is a direct (though certainly unintentional) consequence of the network architecture. By explicitly and directly naming all the potential recipients, we create a system where the senders can easily express their desire to send data to any re- cipient in the network. Given that, under the typical flat-fee contracts, the marginal cost of sending additional packets is very close to zero (up to some capacity limit). Hence, there are few or no incentives for refraining from sending unwanted traffic; sending some more packets, either just for fun in order to gain legitimate or illegiti- mate profits, costs so little that it doesn’t matter. Hence, for SPAM, even a marginal response rate creates a strong incentive for sending unsolicited advertisements, and for DDoS-based extortion, even a small success rate creates a strong incentive to launch attacks. To summarise the current unwanted traffic problem is a compound result from the following factors: • An architectural approach, where each recipient has an explicit name and where each potential sender can send packets to any recipient without the recipient’s consent.
- 34. 4 L.M. Correia et al. • A business structure, where the marginal cost of sending some more packets or messages (up to some usually quite high limit) is very close to zero. • The lack of laws, international treaties, and especially enforcement structures that would allow effective punishment of those engaging in illegal activity in the In- ternet. Basically, the separation of identifiers and locators can be used to create archi- tectures where a sender must acquire the recipients’ consent before it can send any data, beyond some severely rate-limited signalling messages. 1.1.4 Configuration and Management Complexity Networks are becoming larger, more heterogeneous, and more dynamic. End users expect ubiquitous service availability on a variety of devices and equipment. More equipment and network types will coexist in a single network operator’s domain. Security threats will change as networks and services change. Changing business models are likely to require that network elements are able to enforce access control locally and for instance maintain configuration integrity despite allowing access to important resources from different administrative domains. The traditional operator-to-subscriber based business models are being replaced by several other models, such as the user-to-network model. The emerging user- to-user model is a challenging model for operators: Web 2.0 services with user- generated content exist only in the service plane. Typically, end-user services are composite Data-Multimedia-Voice-Video (DMV2) services. In this model, users get access to the content of other users, with revenue generation typically provided by advertising revenue. Outsourcing of management entirely or partially to one or more outsourcing providers is another common business relationship requiring means for tracking service delivery liability. As a consequence of this business innovation, the technical network operations perspective of the traditional operator is being substi- tuted with a business-focused service management perspective, where service de- livery according to end-user expectations is crucial and lower level network aspects are not significantly interesting. The total cost of ownership for service enabling equipment is highly focused by operators. At present, there are typically numerous heterogeneous management displays from different vendors. They do not provide sufficient input to business decisions and prioritisations, which makes deployment and assurance of even a very small service a time-consuming challenge. Being to a large extent based on humans, this management doesn’t scale. From a vendor perspective, the current ad-hoc design of element and network management instrumentation and systems is costly in relation to business value for the service provider. Network element instrumentation consist of hundreds of per- formance counters, events, alarms and configuration parameters. This challenges both network element vendor management system developers and the operations staff.
- 35. 1 Introduction 5 1.1.5 Lack of Privacy and Accountability The aim of privacy and accountability is to prevent socially undesirable things from happening, on one hand by imposing technical restrictions on informa- tion flow, and on the other hand by creating explicit incentives for desirable be- haviour. The privacy problem is a complex one, with at least three different viewpoints. From the Orwellian point of view, the question is about freedom of speech and governmental control. A sufficient privacy system ensures that we can express our opinions and think freely, within reasonable bounds (like not committing clearly criminal acts) even when our opinions are socially unacceptable or hostile to- wards the governing regime. The Kafkaesque aspect of privacy focuses on citizen’s ability to retain their autonomy without fear of unfounded litigation or other ha- rassing legal/other action. Thirdly, the economic aspect of privacy relates to the fine balance between socially beneficial differentiated pricing vs. socially harmful price discrimination. From these three different points of view, it seems a neces- sity to provide a reasonable base-level of privacy as a built-in feature in future net- works. The flip side of privacy is accountability. Unbounded privacy encourages irre- sponsible behaviour patterns, such as rampant advertising. To counter these, in- creased privacy requires increased accountability; a fact that appears as a paradox from the technical point of view. A key to understanding this technical paradox is to consider the different dimensions of communication. At the baseline level, we can make a difference between four dimensions: the content of communication, the parties communicating, their locations, and finally the very fact that a piece of communication took place (existence). If the system is able to provide strong “insu- lation” between these dimensions so that each party gets only the relevant pieces of information, a high level of privacy can be preserved. For example, a communica- tions service provider needs to know that communication takes place and whom to attribute the communication to, but should have no access the content, the identity of the other parties, nor their locations. 1.1.6 Poor Support for Mobility and Multi-homing Effective mobility support requires a level of indirection. It is needed to map the mo- bile entity’s stable name to its dynamic, changing location. Effective multi-homing support (or support for multi-access/multi-presence) requires a similar kind of indi- rection, allowing the unique name of a multi-accessible entity to be mapped to the multitude of locations at which it is reachable. Within the Internet community, the classical approach has been to consider mo- bility and multi-homing as separate, technical problems. The main result of this are the Mobile IP protocols, which are architecturally based on re-using a single name space, the IP address space, for both stable host identifiers (Home Addresses)
- 36. 6 L.M. Correia et al. and dynamic locators (Care-of Addresses). While the approach certainly works, it creates two major drawbacks. At the same time, the tendency of considering multi-homing a separate problem with a separate solution creates feature interac- tions. With regard to the Mobile IP approach, it binds the communication sessions (TCP connections and application state) to the home addresses. This, in turn, when com- bined with the only known scalable solutions to a number of related security prob- lems, creates an undesirable dependency on a constant reachability of the home address. In other words, the Mobile IP architecture is intrinsically bound to the availability of the home addresses; the home agent becomes a new single point of failure. Secondly, approaches that use names from a single name space for multiple pur- poses create a number of potential semantic problems. When Mobile IP is used, there are no easy way to tell if two IP addresses actually point to a single host (e.g., due to one being its home address and another one its care-of address) or not, i.e., whether one is merely an alias for the other or an identifier for a genuinely different node. That, in turn, may lead to very confusing problems for quite a large number of applications. 1.2 Short 4WARD Overview 4WARD performs research on the architecture of a Future Internet adopting a “clean slate” research approach. This means the practical constraint of evolving from the existing TCP/IP-based network architecture is temporarily ignored in the interest of discovering a design that is ideally adapted to present and expected future usage and is not forced to adapt to architectural decisions made some thirty years ago with quite different objectives and constraints. An architecture designed following this approach may be seen as a target for the current network to evolve to. It may alternatively be seen as the blueprint of a parallel architecture that could coexist and interoperate with IP, gradually expanding and taking over the functions of the old network. The strategic objective of 4WARD is to increase the competitiveness of the Eu- ropean networking industry and to improve the quality of life for European citizens by creating a family of dependable and interoperable networks providing direct and ubiquitous access to information. 4WARD’s goal is to make the development of net- works and networked applications faster and easier, leading to both more advanced and more affordable communication services. To achieve this strategic objective, work in 4WARD is guided by 4 overriding tenets: • Tenet 1: Let 1000 Networks Bloom The project explores a new approach to the creation and co-existence of a multi- tude of networks: the best network for each task, each device, each customer, and each technology. 4WARD aims to create a framework in which it will be easy for
- 37. 1 Introduction 7 many networks to bloom as part of a family of interoperable networks that can co-exist and complement each other. • Tenet 2: Let Networks Manage Themselves The 4WARD architecture incorporates an embedded management entity, which is an inseparable part of the network and each of its components, generating extra value in terms of guaranteed performance in a cost effective way, and capable of adjusting itself to different network sizes, configurations, and external conditions under the control of policies set by the network owner. • Tenet 3: Let a Network Path Be an Active Unit A forwarding path is recognised as an active network component that controls itself and provides customised transport services. An active path can provide re- silience and fail-over, offer mobility, simultaneously use multiple different se- quences of links, secure and compress transmitted data, and optimise its perfor- mance. • Tenet 4: Let Networks Be Information-Centric Users are primarily interested in using services and accessing information, not in the nodes that host information or provide services. Consequently, the 4WARD architecture considers information objects (and their digital instantiations) and services as primary importance that are not tied to any particular device but can rather be mobile and distributed throughout the network. Such, 4WARD ad- dresses one of the fundamental flaws of the Internet architecture. The Future Internet will be even more important for society at large than the present network and 4WARD therefore also performs research on the socio- economic and regulatory issues arising from the application of the above tenets. In our approach, we combine on the one hand innovations needed to improve specific aspects of a network architecture, and on the other hand work on a common overall architecture framework that neatly fit these innovations together. This work is structured into six work packages: three of them consider innova- tions for a single network architecture, i.e., Generic Path, In-Network Management and the Network of Information, one work package studies the use of Virtualisation to allow multiple networking architectures to co-exist on the same infrastructure, another work package looks at the design and development of Interoperable Archi- tectures, and finally one work package that ensures that all envisaged developments take proper account of essential Non-Technical Issues. 4WARD is an Integrated Project assembling 36 partners in a strong, industry-led consortium of the leading operators, vendors, SMEs, and research organisations. The consortium includes partners from North America and Asia and has a strong background of research on networking architecture with particular expertise in the field of wireless and mobility. The project has originally been granted a budget of 23 M€ for a period of two years, but it has been extended for another half year to match with future Call 5 projects.
- 38. 8 L.M. Correia et al. 1.3 Position of 4WARD in Europe and EC Projects and Other Regions 1.3.1 EU Framework Programme 7 In the EU Framework Programme 7 (FP7) there are several projects that are relevant for this area. Some of them are having a clean slate approach others are working with an incremental approach to resolve some of the issues in the current Internet within the present paradigm. A couple projects that are relevant for this area are listed below: • PSIRP is a STREP with the aim to investigate the “publish–subscribe” paradigm. • Trilogy is an IP lead by BT to resolve the current problems with BGP within the present paradigm. • Sensei is an IP working with sensor networks and trying to create open service interfaces and corresponding semantic specification to unify the access to con- text information and actuation services offered by the system for services and applications. • Onelab2 is an IP that deals with creation of a testbed for experiments for Future Internet. • Moment is a STREP dedicated to handle bandwidth measurement. The Commission is, in addition, trying to coordinate the Future Internet activi- ties more actively and has also established a Future Internet Assembly amongst the research projects to further the activities on Future Internet and coordinate across a number of domains like content media, security networking, etc. 4WARD has played a prominent role in FIA and have responsibilities as caretakers helping to organise the meetings and sessions, 4WARD has through its coordinator also been driving the Future Internet clusters where a lot of architecture and scenario work has taken place and been used as input also to the Future Internet Assembly. 1.3.2 FIND (Future Internet Design) US The Future Internet Design (FIND) program was initiated by National Science Foundation (NSF) in 2006 with the objective of supporting a wide range of small- to-medium sized “clean-slate” protocol investigations across the academic research community. The scope of the program includes trust, security, impact of emerging wireless and optical technologies, network economics and social aspects. In 2009, NSF organised an external panel review of the FIND program, involv- ing a detailed evaluation of over 30 projects. The panel provided a strong positive recommendation about the program, commenting on the benefits of clean-slate re- search without the usual constraints of backward compatibility with existing net- work protocols. The panellists felt that new ground was being broken on important
- 39. 1 Introduction 9 research topics such as: naming, addressing, routing, monitoring, mobility, network management, access and transport technologies, sensing, content and media deliv- ery, and networked applications. The panel recommended that NSF continue the program and initiate an integrated community effort to build teams who would de- sign and prototype more comprehensive converged future Internet architectures. The panel also recommended an increased focus on security and network management aspects. NSF accepted these conclusions and formed a new program called “Future Internet Architectures (FIA)” (NSF 10-528) that would support 2–4 large project teams working on comprehensive and converged future Internet architectures. These projects are expected to result in a completed design, protocol validation and initial deployment on infrastructures such as GENI. 1.3.3 GENI (Global Environment for Network Innovation) US The GENI (Global Environment for Network Innovation) program was initiated by NSF in 2008 with the objective of developing flexible and large-scale networking infrastructure for future Internet research being done under FIND (Future Internet Design) and other programs. GENI is managed by the GPO (GENI project office) at BBN Technologies, Cambridge, MA and is headed by Chip Elliott, Program Man- ager. The approach adopted by GENI is based on a number of principles including: • Spiral development with continuous improvement and feedback • Leveraging existing capabilities and testbeds across US research community • Federation of testbeds and campus networks to form an integrated GENI facility • Competition among research groups for selection of key GENI components • Open, collaborative project with open-source software, international partners, etc. GENI has been organised into Spirals, with Spiral I starting in Nov 2008 and ending in Nov 2009, and Spiral II starting in Dec 2009 and ending in Dec 2010. The first spiral emphasised technology evaluation and risk mitigation through proof- of-concept prototypes. The second emphasises integration of an initial federated “meso-scale” GENI prototype across ∼8–10 campus locations, with a unified ex- perimental control and management interface. 1.3.4 Akari Japan The objective of the AKARI Architecture Design Project (in short AKARI Project) is to design the network of the future. The AKARI Project aims to implement a new generation network by 2015 by establishing a network architecture and creating a network design based on that architecture. The motto is “a small light (akari in Japanese) in the dark pointing to the future.” The philosophy is to pursue an ideal solution by researching new network architectures from a clean slate, without being
- 40. 10 L.M. Correia et al. impeded by existing constraints. Then the issue of migration from existing networks can be considered. The goal is to create an overarching design of what the entire future network should be. To accomplish this vision of a future network embedded as part of societal infrastructure, each fundamental technology or sub-architecture must be selected and the overall design simplified through integration. The AKARI project schedule is divided into two five-year periods: the first five- year period (FY 2006–2010) aims at finalising the new generation network design blueprint and the second five-year period (FY 2011–2015) will develop test-beds based on the blueprint. In the first year (FY 2006), the conceptual design was cre- ated and initial design principles were presented. Detailed design was performed during the second year while revising the initial design principles. Prototypes will be developed, evaluated, and verified to indicate the validity of the concepts. Design diagrams will be completed in the fifth year in the first five year period. In the sixth and subsequent years, the new generation network concepts will be incorporated in test-beds based on the developed prototypes and design diagrams to conduct demonstration experiments. In addition, the network components will be created and protocol engineering will be performed to establish new generation network construction techniques. 1.4 The Book The book from the 4WARD project deals with Architecture and Design for the Fu- ture Internet and is covering a broad spectrum of issues. We give a system overview and give a socio-economic background reasons and regulations for a Future Internet, but also go into some depth of the different technical issues. Chapter 2 on System Overview describes the System Model and defines the struc- ture and behaviour of a system that is to be constructed as well as its generativity, i.e., how bigger and more complex future systems and networks could be built by using a small set of generic concepts. 4WARD is promoting a new approach to net- working based on the analysis of both the success factors of the Internet (seen as the core Internet design principles and core IP protocols) as well as the factors that have led to ossification and the patchwork type of the IP evolution of recent years. The Network of the Future must be based on a new set of architectural principles. It is well understood that the development path of any industry or economic sec- tor is significantly affected by the opportunities provided by the available technolo- gies, the particular characteristics of its markets and the directions and priorities of related government policies and regulations. Previously, there has been a tendency to leave these issues to be handled separately, and the non-technical topics above have been addressed after the technology had been developed. In the case of the global networked society, this is not a desirable approach. The take-off and suc- cess of the Future Internet will be closely linked with what actions are taken on all areas of the Future Internet ecosystem. Chapter 3 of the book on Socio-economic aspects describes how major non-technical drivers impact the transition from the
- 41. 1 Introduction 11 RD stage to the real deployment of the technical and architectural innovations studied in 4WARD. Virtualisation is a key technology for the deployment of new customised net- work architectures. After a short introduction into the overall concept of network virtualisation, its goals and benefits as well as scenarios and business aspects are presented. Then the virtualisation framework is described in more detail, starting with an overview of the process for building and setting up virtual networks, and in- cluding resource virtualisation, and provisioning, control and management of virtual networks. Afterwards the design process to be followed by the network architect, the design of new network architectures is described. In this process, the network archi- tect can also follow the design patterns described in Chap. 4 on “How to design and build networks” in order to: (i) effectively compose different functionalities to meet the initial requirements, and (ii) assure the interoperability among different architec- tures, taking business relationships, security and management issues into account. In order to assure the interoperability among virtual networks, the concept of folding points is analysed in detail. Naming and addressing has been a source of considerable contention in exist- ing network designs. What precisely is named, what an address is, and how these two concepts relate to each other by name resolution has been treated differently and inconsistently in different systems as well as in different architectures. The 4WARD project pursues an integrated, coherent approach for a naming address- ing architecture that combines flexibility with coherence and integrates its different components via a cross-layer name resolution concept. Chapter 5 on Naming and Addressing will discuss the basic design rationale of this concept. It will also go over some examples, ranging from very simple, local naming/addressing schemes, over schemes intended for consistent naming and addressing in a network layer ex- tending world-wide, to a rather complex naming/addressing structure suitable for a data-centric network of information. All these schemes combine into the over- all naming and name resolution architecture, yet remain flexible at their respective layers of abstraction. Chapter 6 on Security principles gives some considerations for how rethinking the fundamental network architecture affects and is driven by security considera- tions. The information-centric approach of 4WARD is built on the concept of secur- ing information rather than the containers containing information. Doing so, the se- curity 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 has only begun addressing the security principles necessary for dynamical manage- ment of virtualised, 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 largely out of scope. 4WARD acknowledges and considers the business and governmental con- trol interests that will heavily influence the security direction into which the future network evolves.
- 42. 12 L.M. Correia et al. One of the key challenges for the Future Internet is the correct definition and im- plementation of the domain concept. Chapter 7 analyses the interconnection model of the Internet and of current Mobile Operators and presents the inter-domain con- cepts developed in the scope of 4WARD. Special attention is devoted to the still to solve problem of Multi-domain Quality of Service. The cost and complexity of configuring and running networked services are significant and expected to increase. We propose a solution for management, In- Network Management (INM), which is based on decentralisation, self-organisation, and autonomy of management processes. Its key idea is that management stations outside the network delegate management tasks to the network itself, supporting fu- ture large-scale networks that self-configure, dynamically adapt to external events and allow for low-cost operation. In this Chap. 8 on “How to manage networks”, we will 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 monitor- ing in a distributed manner are presented, and self-adaptation schemes for resource control are discussed. Transporting information through the Internet has traditionally been following the end-to-end principle. This means that no knowledge about the nature of the transported information is assumed within the network and leads consequently to overlay networks realising specific services. Keeping state information “in the net- work” is generally seen as a burden for scalability and undesirable. However, mobil- ity of hosts and applications, any guarantees for quality of service, and new methods for cooperation and coding, all require a certain amount of information to be stored at specific places inside the network. Chapter 9 describes an architecture for data transmission that puts technological and administrative domains (compartments) in the role of the keeper of this shared information. Paths are established between communicating entities, basic functional blocks that re-appear 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. Com- partment 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. Chapter 10 presents the overall vision for a network of information, illustrates the fundamental ideas, and explains the mechanisms currently under development that will bring about a major paradigm change in networking. After briefly review- ing relevant scenarios where the current host-centric approach to information stor- age and retrieval is ill-suited for, we introduce how a new networking paradigm emerges, by adopting the information-centric network architecture approach. We il- lustrate how information retrieval may look like in the future, emphasising on the user perspective. We then put forward the architectural requirements for a network of information, and the research directions taken during the project. The core of this chapter centres on a lucid description of the mechanisms, the “nuts and bolts” so
- 43. 1 Introduction 13 to speak, of the technologies that implement a network of information. We describe a network of information operation, providing concrete examples and highlighting the performance improvement expected to materialise with the deployment of a net- work of information. Finally, we take a long-term view and discuss how a network of information can evolve. This chapter concludes with a comprehensive summary of the main network of information innovations and future items of work. In the preceding chapters, we have described concepts and technologies that can be used for designing and building networks, how networks can be intercon- nected and be managed, how connectivity can be established, and how to manage and search for information objects. In addition, we have also established important security principles and schemes for naming and addressing. Together, these provide a foundation and a set of tools for new ways of networking in the Future Internet. In order to show their advantages compared to current paradigms in networking, as well as to show how they can be applied in a consistent and coherent manner, Chap. 11 on Use Cases describes through a set of use cases how complete and in- tegrated solutions for networking can be provided using the principles and tools described in chapters above. They will take us all the way from the design of suited network and software architectures, further on to describe how functionality and interfaces are being deployed, and finally how this functionality is being used and managed in order to carry out the specific tasks described by each of the use cases. To support the theoretical ideas developed within 4WARD, some of them have been realised as prototypes. The experiences collected while implementing the dif- ferent concepts gave valuable feedback and enhanced the ideas with crucial details. The most important concepts have been successfully tested, and this chapter will give an overview of the developed prototypes. Some of them are also publicly avail- able. Pointers to the releases are given in the respective sections of Chap. 12 on Prototype Implementation. The final Chap. 13 gives some conclusions and also describes some migration approaches by the 4WARD project to make it possible to realise the research find- ings.
- 44. Chapter 2 A System Overview Martin Johnsson Abstract The 4WARD System Model is described, defining the structure and be- havior of a communication system that is to be constructed as well as its genera- tivity, 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. 2.1 Background and Motivation This section describes the 4WARD System Model, which defines the structure and behavior of a communication system that is to be constructed as well as its gener- ativity, i.e., how bigger and more complex future systems and networks could be built by using a small set of generic concepts. Through 4WARD, a new approach to networking based on the analysis of both the success factors of the Internet (seen as the core Internet design principles and core IP protocols) as well as the factors that have led to ossification and the patch- work type of the IP evolution of recent years has been developed. The Network of the Future must be based on a new set of Internetworking prin- ciples. These principles are characterized below as four programmatic tenets: M. Johnsson () Ericsson Research, Stockholm, Sweden L.M. Correia et al. (eds.), Architecture and Design for the Future Internet, Signals and Communication Technology, DOI 10.1007/978-90-481-9346-2_2, © Springer Science+Business Media B.V. 2011 15
- 45. 16 M. Johnsson 1. Let 1000 Networks Bloom We will explore a new approach to a multitude of networks: the best network for each task, each device, each customer, and each technology. Unlike the multi- tude we had in the past, where different incompatible technologies were compet- ing with each other, we want to create a framework that will allow these many networks to bloom as a family of interoperable networks coexisting and comple- menting each other. 2. Let Networks Manage Themselves The main limits of current technologies are the scaling up to very large network sizes, and the needed human intervention which is associated with considerable cost, errors and with an inherent slowness in reacting to changing network condi- tions. What we would like to have is a management entity as an inseparable part of the network itself, generating extra value in terms of guaranteed performance in a cost effective way, and capable of adjusting itself to different network sizes, configuration, and external conditions. 3. Let a Network Path Be an Active Unit We want to consider a path as an active part of the network that controls itself and provides customized transport services. An active path can provide resilience and fail-over, offer mobility, simultaneously use multiple different sequences of links, secure and compress transmitted data, and optimize its performance all by itself. 4. Let Networks Be Information-Centric Users are primarily interested in using services and accessing information, not in accessing nodes that host information or provide services. Consequently, we want to build a network as a network of information and services that may be mobile and distributed. In such a network, the users just accesses items of interest by their name while the data locations can be completely hidden. These tenets, together with the understanding of the current situation of today’s Internet, formed the main drivers for the definition of the 4WARD Technical Re- quirements [1], which laid the foundation for technical work within the 4WARD project. This work ultimately resulted in the 4WARD System Model, which is de- scribed in the following section. 2.2 The 4WARD System Model Figure 2.1 depicts the 4WARD System Model, which has been developed with the Tenets and the 4WARD Technical Requirements [1] as main principal input. The system model gives the necessary definitions, specifications, principles and guide- lines for designing, building, deploying, and manage interoperable network archi- tectures. For that purpose, the 4WARD System Model consists of an Architecture Framework and a set of Architecture Pillars which provides the essential technolo- gies in many of the network architectures anticipated and required for the future networks, though it is possible to also deploy and use them in migration scenarios. With the 4WARD System Model we expect significant efficiency gains in the de-
- 46. 2 A System Overview 17 Fig. 2.1 The 4WARD System Model sign, management and operation of networks, which is one of the key challenges in both current and future networks. The Architecture Pillars have been defined using a new set of concepts and technologies to address emerging business models and new types of applications: • A new abstraction and model of the physical and virtualized infrastructure, in- cluding all of transmission, processing, and storage resources. • ONE modular and extensible connectivity concept, supporting all modes and topologies of endpoint associations. • A new open model and API for content and information management. Search and retrieval of information objects using a persistent identity. • Management providing an inherent capability of the functions in the network. The Architecture Framework provides a unified component-based design pro- cess, 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 design process includes the architectural principles and re-usable design pat- terns at various levels of abstractions out of which families of interoperable network architectures can be defined. The Architecture Pillars: In-Network Domain Management, NetInf, Generic Path, and the Physical but virtualized substrate (and each of them in turn define their own respective frameworks or architectures) themselves to be defined by using the Architecture Framework.
- 47. 18 M. Johnsson The Physical Substrate provides an abstraction of the physical resources of any network spanning from the smallest to the largest. The abstraction is the key for co- herent virtualization and management of those underlying resources across domain borders. The result of a virtualization operation is a virtualized network, providing resources onto which an operator is free to instantiate its own choice of functions, protocols, etc., for example Generic Paths and NetInf. The Generic Path provides a generalized transport mechanism to transfer data between entities in the network. The recursive Generic Path concept is able to model virtually any type and level of transport, be it point-to-point or multipoint- to-multipoint, or supporting transport on links at the physical level, or end-to-end across networks. Generic Paths specifically give support for the dissemination of information objects. NetInf (Network of Information) provides for identification, management, and dissemination of information objects. NetInf is a new abstraction of information (and service) management, where applications do not need to be aware of where an information object is stored. In-Network Management (INM) is omni-present in all network functionalities. It provides design patterns and interfaces as well as more specific mechanisms, fa- cilitating various degrees of self-management capabilities. This spans from such capabilities living ‘beside’ the functionality it is supposed to manage, and then all through to functionalities being fully and inherently self-managed. A special case of In-Network Management is In-Network Domain Management, which provides self-management capabilities on domain as well as inter-domain scale. The Knowledge function (also known as Knowledge stratum) discovers, gath- ers and further infers status of network topologies, resource and context status by querying the network functionalities operating in the network, for example Generic Paths and NetInf. The Governance function (also known as Governance stratum) provides control and management of network functionalities, and governs by query- ing the status of the network from the Knowledge function. The Governance func- tion will decide out from policies (provided by a network administrator) as well as the network status what network functionalities shall operate in the network. Gov- ernance and Knowledge functions are also instrumental for the interconnection and composition of networks and domains, where dynamic and highly automized cre- ation of SLAs is supported. The following sections provide an overview and introduction of the concepts and technologies that make up the foundation of the Architecture Pillars, and it serves as an introduction to the contents provided through Chap. 4–10. 2.3 The Architecture Framework 2.3.1 Strata, Netlets, and the Design Repository The Architecture Framework must provide ways to (i) guide the Network Architect to allocate the required network functionalities and (ii) assure the interoperability within families of network architectures.
- 48. 2 A System Overview 19 Fig. 2.2 High-level view of 4WARD Architecture Framework As can be seen in Fig. 2.2, the following main components constitute this frame- work (see Chap. 4 for further detail): • A Stratum is modelled as a set of logical Nodes which are connected through a Medium that provides the means for communication between the Nodes in- side this stratum. This stratum encapsulates functions that are distributed over the nodes. These functions are provided to other strata through two well known in- terfaces (that can be also distributed over the nodes): The SSP (Service Stratum Point) that provides the services to the other strata located on top of the respective Stratum and to the vertical strata. Figure 2.2 shows Stratum Y using the services provided by Stratum X through SSPX. The SGP (Service Gateway Point) offers peering relations to other strata of the same type. • Strata can manage themselves. For example, when a routing service stratum is deployed, it organizes itself onto the physical infrastructure. The deployment will be in accordance with the specification of the logical nodes and the medium of the stratum, taking then into account the topology, capabilities, and resource status of the nodes and links in the physical infrastructure. • Horizontally stacked strata (as shown in the middle of Fig. 2.2) are related to the transport and management of data across networks. Within such strata, Netlets can be considered as containers for networking services. They consist of func- tions/protocols inside a Node that are needed to provide the services. By virtue of containing protocols, Netlets can provide the Medium for different Strata, i.e. inside the same Netlet there could be functionalities that are related to different strata. Figure 2.2 shows such Netlets implementing media for different strata in- side the same node. • The two vertically oriented strata provide Governance and Knowledge for an en- tire network (i.e. a set of horizontal strata). The Knowledge Stratum provides and maintains a topology database as well as context and resource allocation status as reported by a horizontal stratum. The Governance Stratum uses this information, together with input provided via policies, to continuously determine an optimal
- 49. 20 M. Johnsson configuration of horizontal strata to meet the performance criteria for a network. The Governance Stratum also establishes and maintains relations and agreements with other networks. The Repository contains the set of Building Blocks and Design Patterns for the composition of functionalities (i.e., to construct the strata and the netlets) for specific network architectures, including best practices and constraints to ensure interoper- ability between network architectures. 2.3.2 The Design Process Evolution of today’s networks including the Internet suffers from the inability to be extended in a consistent and reliable way while maintaining certain assured prop- erties, such as security, quality of service, reliability even in the broader context. Much effort has to be spent for standardization, development and regression testing when introducing even minor feature improvements, before deploying them on a network-wide basis. Upgrading of a large installed base of network elements means a big technological challenge and financial risk to the network operator and service provider. 4WARD has succeeded in setting up a design process that in the future will enable new network designs to be developed, tested and deployed without impacting the installed network basis, when based on this 4WARD architecture framework and building upon the recent progress in network virtualization. The innovative 4WARD network design process leverages advantages of model-driven software engineering techniques and the experiences in design and composition of web services, based on OSGI principles [2]. As shown in Fig. 2.3, the following phases are considered in the design process: 1. Requirements Analysis: Starting from the business idea and requirements, the goal of this step is to decompose them into the high level functionalities that should be realized by the architecture to be designed. The output of this phase is mainly the identification of the macroscopic architectural view of Strata, a first draft of the main network components, and the specification of technical require- ments for further refinement of the architecture. 2. Abstract Service Design: During this phase, the technical requirements and the high level functionalities derived from these will be turned into abstract func- tionalities and ways how they can be composed, following generic principles and design patterns. The result of this design phase is the specification of the Netlets operating at node level, and the Strata that constitute the distribution of functionalities across the network nodes. 3. The Component Design Phase focuses on the detailed specification and com- position of the Functional Blocks (FBs) used to implement the specific function- ality. This includes the specification of the interfaces, properties, and require- ments/prerequisites of the FBs. The output of this phase is the detailed design of
- 50. 2 A System Overview 21 Fig. 2.3 High-level view of the 4WARD Design Process the Netlets and software Components, which finally constitutes an “architectural blueprint” ready for instantiation on a network virtualization platform. The entire design process is supported by an integrated design environment, which easily supports backtracking in iterative loops to redesign and improve the results of previous phases. In order to increase the reuse of architectural constructs and store the expertise and knowledge of the designing architect, an “architectural design repository” is used, which contains pre-built architectural constructs (abstract strata, netlets, components, functional blocks) as well as their derived instantiations, proven architectural design patterns on service and network composition, interoper- ability, security, etc. 2.4 In-Network Management INM specifies two key architectural elements in order to realize distributed man- agement within and across the network nodes: Management Capabilities (MC) and Self Managing Entities (SE). The MCs are encapsulations of management logic. The SEs are associated with a specific service and include relevant MCs for man- agement of the service. Both elements are central to achieve autonomous behav- ior. As part of the INM solution and design, algorithms have been developed for real-time monitoring, anomaly detection, situation awareness, and self-adaptation
- 51. 22 M. Johnsson Fig. 2.4 INM relationship with Governance/Knowledge schemes. The MC architectural element is the enabler of these algorithms. These algorithms provide best of breed mechanisms and patterns to address manage- ment tasks. They become important building blocks when designing networks. The 4WARD design process as described above includes an ‘architectural design repos- itory’ which houses design patterns and network type building blocks available to the architect of the future networks. From a management perspective the algorithms developed for INM are key components of this repository which the architect can deploy as the need arises. The ‘management by objective’ approach of INM is intrinsic to governance of networks and knowledge generation inside networks of the future. Both governance and knowledge are modelled as strata in the 4WARD architectural framework. Fig- ure 2.4 shows management objectives being pushed downwards through the gov- ernance stratum, into the SEs and eventually into multiple MCs which carry out the tasks in hand. The MCs in the figure could for example implement a mon- itoring algorithm. The output of the monitoring algorithm is in essence unpro- cessed data. This is fed into the knowledge stratum and reasoned upon and more high level knowledge generated. This knowledge is then used, possibly fed back into governance if some modifications or tweaking are necessary or displayed at a higher level as feedback on the objectives which an operator applied to the net- work. The algorithms developed and the management by objective approach which INM provides are key enablers in the realization of self managing, interoperable networks of the future.
- 52. 2 A System Overview 23 Fig. 2.5 Virtualization ecosystem 2.5 Network Virtualization Virtualization has by now gained sufficient momentum as one of the key paradigms for future networking, as it has the potential to resolve the so-called “deployment stalemate” observed in today’s Internet and foster the development of future net- works paradigms. The straightforward use case for network virtualization is the scenario based on the decoupling of infrastructure ownership and virtual network operation. The virtualization ecosystem encompasses three basic roles, namely (a) the in- frastructure provider (having the capability to virtualize the physical infrastructure by partitioning them into ‘slices’), (b) the virtual network provider (making the pro- visioning of complete end-to-end VNets by putting together ‘slices’ from the un- derlying infrastructure), and (c) the virtual network operator who is operating and managing a VNET. This is illustrated by Fig. 2.5. A service provider is then able to run specific services and applications on this VNet, which are then offered to end users. Communication means between these actors and the definition of the respective interfaces constitute a cornerstone of the network virtualization architecture. This requires the specification of a formal virtual network description, allowing for flexi- bility, extensibility, scalability, interoperability and security. Since multiple business scenarios can be defined (ranging from vertical integration to a strict separation of roles), which imply different relationships of trust between them, the capability to define different levels of abstraction is also a key requirement. The 4WARD Re- source Description Framework provides a language to describe virtual network re-
- 53. 24 M. Johnsson sources and topologies, including all possible constraints that might be applicable in each case. An object-oriented data model was defined with four basic classes describing specific network elements, namely nodes, links, interfaces, and paths. 4WARD network virtualization architecture breaks with the traditional clear sep- aration between a “dumb” core and a feature-rich edge in service provider networks. In this scenario, scalability will be a major challenge, particularly in terms of provi- sioning, management and control of virtual networks. A framework and algorithms for scalable mapping and embedding of virtual resources into the infrastructure, including discovery, matching, and binding were developed. Initial results suggest that the efficient construction of virtual networks from shared infrastructure at large scale is indeed feasible. One of the most important features of the current Internet, global reachability and inter-networking, will surely remain a requirement in the future. This means that vir- tual networks, which by definition are separated and isolated from each other, will still need to communicate, although in a more controlled way. A concept for facili- ties to provide interworking between virtual networks, the Folding Points, has been developed, including the basic elements (Folding Nodes and Folding Links), as well as mechanisms for deployment using the virtual network provisioning framework. 2.6 Generic Paths New mechanisms for data transport face contradictory requirements: large flexibility vs. uniform interfaces to all transport entities and efficient reuse of functionality are required. This can be partially achieved by new protocols only in end systems, but in general, an approach how to structure protocols both at the edge and in the core, at various “layers” is needed. For example, network management needs to identify, inside the network, data flows of different types; they should be able to give account of themselves (e.g., about their desired data rate) and obey a common set of commands. To support such requirements, we focus on the data flow and its path as a core abstraction, along with a design process for a variety of path/flow behaviors. This process can incorporate new networking ideas; examples are network coding, spatial diversity cooperation, or multi-layer routing and is suitable for both end system and in-network implementation; the deployment is supported by the Architecture Framework. The starting point for the 4WARD transport architecture was to find (1) a de- velopment model that can support reuse and flexibility, (2) a proper execution en- vironment within a node (end system or router) with naming and addressing struc- ture and a resolution scheme, and (3) the core functions and APIs necessary for a path, as generic as possible. Together, this is the core of the Generic Path ar- chitecture. It approaches issue (1) by using an object-oriented approach to define types of Generic Paths and to structure their interfaces; issue (2) by defining a set of constructs (namely, entity, endpoint, mediation point, compartment, hooks, and
- 54. 2 A System Overview 25 path) that describe the execution environment of instances of such path types; and is- sue (3) by selecting which operations should be possible on such paths (e.g., joining, splicing, or multiplexing). The concept shares some commonalities with OpenFlow, but concentrates on real-world necessities rather than on experimental usage; it also goes beyond merely modifying switching tables. To incorporate new networking ideas, all the relevant flows in a network share crucial commonalities and provide a common set of APIs with which to manipulate these flows. 4WARD’s “Cooperation Coding Framework” exploits such commonalities by addressing an entity that detects opportunities for turning on cooperation opportunities, like network coding, and can create the necessary path instances to setup a network coding butterfly. Mo- bility may be supported at different levels or compartments—and the realization of mobility at a session level is quite different from the realization of mobility at IP level, though they still share commonalities that can be defined through generalized mobility schemes. Thus, the GP framework allows the abstract description of a mo- bility process in terms of GP constructs, namely, entity, compartment, ports, path, and mediation point. Its realization can then resort to specific technologies adequate to the compartment we are considering in each case. Based on this mindset, it becomes possible to develop powerful, custom-tailored path types. An example are path types for a Network of Information (described next), where the download of documents and the updating of location/caching ta- bles can be tightly integrated and can access topology information to choose, for a document of interest, topologically close caches. Another example would be a path type to support the exchange of management information for In-Network Man- agement entities, e.g., by compressing monitoring information more and more the further it is away from its source. 2.7 Network of Information Today’s networking is essentially about exchanging information between nodes. When accessing information, the request typically includes the host where the in- formation shall be retrieved from, frequently in the form of a Uniform Resource Locator. This host-centric approach is often an obstacle for optimized transport of and easy access to information. Our approach to an information-centric architec- ture puts the information itself on the center stage. We take existing proposals that separate the host identity from the locator one step further by introducing informa- tion objects as first order elements in the network. In addition to classical scenarios such as content distribution, our work also encompasses scenarios that have so far not been discussed in the research community, e.g., the notion of real-world object tracking under the aegis of an information-centric architecture. For the envisaged Network of Information (NetInf), we have developed an in- formation model that constitutes a versatile and widely applicable framework for representing information in a wide sense. A clear split between the information it- self and the location where it is stored is introduced. This eliminates the need for overloading locators and avoids putting them in the role of being an identifier and a
- 55. Random documents with unrelated content Scribd suggests to you:
- 56. Lähtekäämme edes etsimään, hän huudahti ja tarttui Hildegardin käsivarteen; ehkäpä se vielä on jossain salin lattialla. Minä katsastan keskustaa, herrain jalkoihin, tarkasta sinä naisten tuolien alle. Yhdessä he lähtivät kiertelemään salia, näköjään vain huolettomalla huvikävelyllä, mutta sydämessä hornan tuskat, näköjään hilpeästi, rupatellen keskenään, mutta sisällisesti itkien ja käsiä väännellen. Joskus kävi jokin pitkäpiimäinen herra heitä puhuttelemaan ja pakotti heidät pysähtymään ja vastailemaan, vaikka heidän päätään pyörrytti koko ajan ajatus: Nyt sen ehkä joku toinen löytää! Nyt he tulivat vihreitten parven keskeen, siellä heidän täytyi jälleen pysähtyä ja hymyillä joka taholle ja leperrellä mielettömiä kohteliaisuuksia; nyt he saapuivat Eugenie Baldereckin kohdalle, joka kysyi heiltä, eikö pitäisi lisätä vielä joku tanssinumero illan ohjelmaan, ja heidän oli pakko ajatella punakantista kirjaa, jossa oli muuan sattuvasti osattu muotokuva ja sen alla kirjoitus: Nenäkäs, tunteeton ja häikäilemätön on E… B…. He joutuivat — auta armias taivas! — Finkin vaaralliseen läheisyyteen, tuon herran, josta kirjassa oli hirveä kuva, kuinka hän herra von Tönnchenin kanssa istui samalla viiniköynnöksen oksalla, ja kuvan alla säkeet: Vihreävarpu ja pähkinäpihti istuit yhdessä ryyppäämässä; Kärkevyyttänsä varpunen kiitti sekä vihreitä höyheniänsä; pähkinäpihti se syvään huokas: ontto mä oon, että julmaa vallan, mitähän tuokin tietää oikein? Kahdesti he siten kiersivät salin; kolmatta kertaa he eivät enää uskaltaneet lähteä, koska eivät edellisilläkään kerroilla olleet mitään löytäneet. Aivan lohduttomina he palasivat Theonen luo.
- 57. Nyt ei ole muuta kuin yksi keino, sanoi Lenore. Missä herra Wohlfart on? Hildegard pidätti häntä käsivarresta. Ethän vain tahdo kellekään herralle… Minä menen hänestä takuuseen, sanoi Lenore ylpeästi. Hän on uskollinen ja luotettava; missä hän nyt onkaan? Tuolla hän puhelee rouva von Baldereckin kanssa. Molemmat etsijät lähtivät verkalleen salin poikki Antonin luo. Hän seisoi tosin selkä heihin päin, mutta tyttöjen lähetessä hän tunsi vastustamatonta halua silmätä soittajiin. Hän kääntyi ympäri, Lenore seisoi hänen edessään ja iski hänelle merkitsevästi silmää; hän lopetti puhelunsa rouva von Baldereckin kanssa ja omistautui kokonaan neitosille. Herra Wohlfart, Theone Lara on täällä salissa kadottanut pienen punasilkkisiin kansiin sidotun kirjan, noin ison vain… Meidän on kovin tärkeätä saada se heti takaisin; pyydän, olkaa niin kiltti ja etsikää se ja tuokaa heti meille. Onko se painettu kirja? Ei, vaan kirjoitettu, ettekä te saa edes vilkaistakaan sisään, sillä siinä on meidän salaisuuksiamme. Vannokaa minulle, ettette löydettyänne avaa sitä lainkaan. Sen vannon, vastasi Anton juhlallisesti. Kiitän teitä, olkaa vain hyvin varovainen. Anton kiiruhti tungoksen läpi ja kulutti seuraavan neljännestunnin läpikotaisella etsiskelemisellä. Lattialta ei löytynyt mitään, tuoleilta ei löytynyt mitään, palvelijoista ei kukaan ollut löytänyt mitään; kirja oli
- 58. kadonnut jäljettömiin. Sydän täynnä sääliä ja myötätuntoa hän toi neitosille tuon surullisen sanoman. Tanssi alkoi jälleen. Theone ei jaksanut päänkivulta nousta tuoliltaan; hänen sydämensä sisimmän sopukan ovi oli tempaistu selki selälleen, sen sisällys paiskattu kadulle, kaikki hänen tunteensa värjöttelivät alastomina jokaisen katseltavina, kaikki hänen salaisuutensa olivat alttiina raa'an ulkomaailman ivalle. Lenore tunsi kovanonnen iskun pikemminkin puolueonnettomuutena. Ruskeat olivat vaarassa kärsiä tappion, josta eivät kykenisi enää milloinkaan nousemaan. Ja nytkö käydä tanssiin! Tanssimaan tulivuoren päällä, permantona kiehuva laavavirta, ja purkausta voi joka hetki odottaa. Mitä kauemmin liittolaiset pohtivat tilannetta, sitä hirvittävämmältä se heistä näytti; sillä yhäti muistui heille mieleen uusia pikku ilkeyksiä, joita kirjan lehdille oli siroteltu. Kun tanssi oli päättynyt, sattui Fink käymään Hildegardin ohi, hän pysähtyi tämän eteen, kolkutti kenkänsä kärjellä permantoa ja sanoi tyttöön päin kääntyen: Tämä lattia kajahtaa niin ontolta; mitähän tuokin tietää oikein; ehkä jalkaimme alla on jokin kadotettu aarre. Hildegard syöksähti Lenoren ja sairaan sisarsirkkusen luo ja huusi miltei hengettömänä: Herra von Fink tietää koko asian! Ruskeat nauhat rupesivat väpättämään kuin myrskyssä, kolme tytönpäätä painui yhteen ja piti neuvottelua. Vihdoin tultiin siihen tulokseen, että Finkin huomautus oli kyllä sangen hätäännyttävä, mutta täydestä onnettomuudesta ei silti voitu vielä olla aivan varmat. Mutta tämäkin heikko oljenkorsi tuli poistemmatuksi, kun Finkin käytös kävi yhä merkillisemmäksi. Hän laiminlöi tänään kokonaan oman puolueensa, hän puhutteli kaikkia ruskeita, hän istahti Theonen viereen, joka oli ennättänyt jo kolmesti kokea Julian myrkkymaljan katkeruuden ja Capulettin huoneen kukistumisen, eikä
- 59. enää kyennyt pidättämään kyyneleitään. Fink rupesi juttelemaan hänen kanssaan ja pakotti hänet vastailemaan, surkutteli hänen kalpeuttaan ja moitti salin ahdasta ilmaa. Hän kiusasi tyttöparkaa, niin että tämä viimein oli mennä pyörryksiin, ja päätti vihdoin pirullisen kostonsa siten, että osotti Theonelle Hulda Werneria ja kysyi: Mitä pidätte tuosta vihreästä hameesta? Eikö hän siinä näytä kuin viheriävarpuselta? Hänen seuraava uhrinsa oli Lenore. Tämä istui uskollistensa keskellä yhäti ylväänä kuin ruhtinatar, vaikkakin valtikkansa kadottanut ruhtinatar. Koko ruskean parven kuullen kävi Fink häntä puhuttelemaan. Neitonen oli hänelle huomaavaisempi kuin koskaan ennen eläessään; hän rutisti nenäliinansa kokoon, niin että pitsireunus repesi, kyetäkseen levollisesti vastaamaan toisen hymyilyyn. Kaikki kävikin hyvin aina siihen saakka, kunnes Fink huusi ohimenevälle herra von Tönnchenille kesken puhettaan: Hei, Bruno, onko pähkinäin särkeminen lempityötäsi? Benno Tönnchen, joka myöskin kuului vihreihin, katsahti ihmeissään seurueeseen ja vastasi: Eipä juuri; jos neiti Lenore on antanut meille pähkinän särjettäväksi, niin pelkäänpä sen olevan liian kovan minun hampailleni. Sillä oli asia ratkaistu; nyt ei voinut enää epäilläkään; kirja oli Finkillä. Ruskeat nauhat lehahtelivat hätäytyneinä hajalleen, koko puolue näytti kananpoikaparvelta, jonka keskeen haukka on iskenyt. Lenore yksin kokosi väkisin kaiken mielenmalttinsa ja kävi rohkeasti Finkin kimppuun. Te olette löytänyt tuon kirjan, herra Fink. Eräs ystäväni on sen pudottanut ja on hyvin onneton tappiostaan. Sen sisällys ei ole tarkoitettu syrjäisten silmille, ja se voisi tässä seurassa
- 60. aiheuttaa suurta pahennusta. Pyydän, että annatte kirjan minulle takaisin. Kirjan? toisti Fink uteliaasti, minkä ihmeen kirjan? Älkää teeskennelkö, sanoi Lenore, onhan päivänselvää, että se on teillä. En voi uskoa että te kaiken sen jälkeen, mitä olen sanonut teille tukalista seurauksista, voitte silmänräpäyksenkään ajan pitää sitä hallussanne. Minä voisin pitää, nyökkäsi Fink. Te olette aivan liian armollinen, kun oletatte minut niin hienotunteiseksi. Sehän olisi perin julkeata, huudahti Lenore. Mitä suurimmalla mielihalulla olisin niin julkea, jos kirja vain olisi minulla. Kirjaa, joka kuuluu teille tai jollekin ystävällenne ja joka mahdollisesti sisältää jotain teidän kädestänne lähtenyttä tahi teille omistettua, en millään muotoa jättäisi takaisin, jos sen sattuisin löytämään; ja jos saan tietää missä se on, niin en haikaile varastaakaan sitä. Ja kun se kerran on käsissäni, opettelen siitä joka rivin ulkoa. Tulenpa tosiaankin koettamaan saavuttaa mielisuosionne esittämällä siitä teille palasia, aina kuin minulla on ilo tavata teitä. Lenore astui askeleen lähemmäksi herjaajaa, ja hänen silmänsä säkenöitsivät. Jos sen teette, herra von Fink, hän huudahti, niin kohtelen teitä kuten kunniatonta ihmistä ainakin. Fink nyökkäsi hänelle säveästi. Tuo tulistuminen pukee teitä ihmeen hyvin, armollinen neiti; mutta kuinka voittekaan vaatia kunnollisuutta sellaiselta hilpeältä linnulta kuin minä olen? Luonto on jakanut lahjojaan eri tavalla, monelle se on antanut kyvyn sepittää
- 61. runoja, toiset osaavat piirtää pisteliäitä pikku kuvia; minä olen siltä saanut kärkevän nokan, jota minun on kaikella ahkeruudella käyttäminen. Oletteko koskaan nähnyt kunnollista viheriä varpusta? Hän kääntyi nauraen poispäin, työnsi kätensä Benno Tönnchenin kainaloon ja lähti tämän kanssa ovea kohti. Lenore riensi Antonin luo. Kirja on herra von Finkillä! Minä rukoilen teitä hankkimaan sen meille takaisin; itse hän kieltäytyi sitä luovuttamasta. Hän ei saa sitä enää lukea, se olisi Theonen kuolemaksi. Anton sieppasi sukkelasti päällystakkinsa ja juoksi ystävänsä perään, joka jo odotteli kadulla. Feroniin, Anton! huudahti Fink hänelle Benno Tönnchenn rinnalta. Minun täytyy saada puhella kanssasi kahdenkesken, sanoi Anton, käyden hänen toiselle kupeelleen. Etpäs nyt vain, sinä ruskea lähettiläs, huudahti Fink nauraen, nyt en tahdo olla missään tekemisissä sinun kanssasi. Kuule, Fritz, minä pyydän sinua, ahdisti Anton häntä, painautuen häneen kiinni, anna tuo kirja takaisin. Tytöt hätäilevät sen takia aivan menehtyäkseen. Anna heidän hätäillä! sanoi Fink. Kukaan heistä ei ummista tänä yönä silmiään. Sen parempi, emme mekään tahdo ummistaa. Hehän voivat tulla joukolla Feroniin, jos heillä kotona käy liian tukalaksi. Me istumme siellä aamuun saakka. Ja sinä, Anton, et nyt pääsekään pujahtamaan
- 62. kotia ilman minua vaan saat kestää sen kuin minäkin ja koko ajan hiljaisessa kuolemantuskassa. Mikä juttu tuo pakina kirjasta on? kysyi Tönnchen Finkin toiselta kupeelta. Älä sano, pyysi Anton hiljaa. Onpahan vain koko hullu juttu, vastasi Fink, saatte kohta kuulla kaikkityyni. Herran tähden, vaikene! hätäili Anton. Minun käytökseni tulee riippumaan sinun käytöksestäsi, sanoi Fink. Jos juokset kesken tiehesi, niin luen toisille kirjan kannesta kanteen. He saapuivat Feronin viinitupaan. Anton punnitsi mielessään, pitikö hänen ehkä hyökätä Finkin kimppuun ja ottaa häneltä kirja väkivalloin. Mutta hän oli epävarma menestyksestä. Vakavuudella ja pyytelyillä ei tänä iltana myöskään saanut mitään aikaan. Vain kavaluuteen voi turvautua. Hänen näitä miettiessään herrat keräytyivät pieneen takahuoneeseen, joka oli heidän tavallinen juomatupansa. Paitsi Antonia ja Finkiä olivat saapuvilla Zernitz ja Tönnchen, pikku Lanzau, eräs Wernereitä, eräs Baldereckien serkku (tämä mulkosilmäinen nuori herra oli kirjassa mainittu lehtisammakon nimellä), kaksi Tronkaa, ei Tronka-Hamsin haaraa vaan toista, majoraattihaaraa. Mitä nyt juodaan? kysyi Fink. Joka miehelle oma pullonsa, ehdotti Zernitz. Ka, miksikäs ei! huudahti Fink.
- 63. Mutta ei vain sitä teidän hirveätä valkoista Burgundiviiniänne, huomautti Guido Tronka. Vielä tänäänkin ovat suoneni viime istunnostamme lähtien kireinä kuin viulunjänteet. — Siis sektiä [saksalainen kuohuviini, sampanjaa muistuttava] ja portteria, rehellisesti puoleksi kumpiakin, ehdotti Fink. Se kelpaa! huudahti pikku Lanzau… Samanlaista hornanjuomaa sekin, valitti Zernitz. Kellarimestari, viinuri, edeskäypä hoi! huusivat herrat ja tekivät tilauksensa. Tällävälin Anton keksi epätoivoisen keinon. Hän pistäytyi huoneesta ulkona, antoi tarjoilijalle taalerin ja pyysi häntä lämmittämään pikku takahuoneen hehkuvan kuumaksi, vähääkään välittämättä toisten herrojen vastalauseesta. Itse hän istuutui niin kauas uunista kuin mahdollista ja näki suureksi ilokseen Finkin painautuvan aivan kiinni sen rautalieriöön. Piankin täytyi helteen käydä hänelle rasittavaksi; silloin heittäisi hän takin päältään, kuten hänen tapansa oli sellaisissa tilanteissa, ja sitten kävisi Antonille ehkä mahdolliseksi ottaa punakantinen kirja takintaskusta Finkin itsensä nähden. Minullapa on teille suuri uutinen kerrottavana, alotti Tönnchen. Oletteko nähnyt Tronkan Alicea, Fink? Enkä, sanoi Fink, kaataen lasiinsa; onko se hevonen vai joku naikkonen? Hevonen tietystikin! huusi Tönnchen. Pah, jättäkää tallitakki tänään kotiin, virkkoi Fink. Mutta tämä on vietävän totta! huudahti
- 64. Tönnchen. Guido on ilmoittautunut sillä herrain kilparatsastukseen. Maksakaa sitten vain purkajaisrahat, sanoi Fink Guido Tronkalle, ja jääkää kauniisti kotiin. Ajaxia ei lyö laudalta mikään juoksija tässä maailmannurkassa. Tulkaahan huomenna katsomaan minun Aliceani, pyysi Tronka hartaasti. Tahtoisin kuulla teidän arvostelunne siitä. Oletteko nähnyt teatterin uutta ensimmäistä rakastajatarta? kysyi Zernitz Antonilta. Hänellä on kauniit silmät. Sillä on mainio päänasento, huusi toinen Tronka pöydän yli Finkille. Hänellähän on ristihuuli, pisti lehtisammakko ylenkatseellisesti väliin. Mitä siellä nyt taas jauhetaan? kysyi Fink. Puhumme Seppistä, tuosta viheriäsilmäisestä syöjättärestä, älähti jälleen sammakko- Baldereck. Ettekö te enää ollenkaan käy teatterissa? En, vastasi Fink, mutta minä lähetän sinne tallirenkini. Jos herrat tarvitsevat tukevakouraista tapattajaa, niin kääntykää vain hänen puoleensa. Alkoi tulla aika lämmin. Anton katsoi parhaaksi kääntää toisten huomion uunista toisaalle. Hän pyysi herra von Zernitziä esittämään erään leikillisen murrejutun, jonka luutnantti oli äskettäin kertonut hänelle itselleen; hän yhtyi äänekkäästi lehtisammakon naurunremahdukseen ja viekoitteli vanhemman Tronkan juttelemaan erään metsästysseikkailun, jonka traagillisena päätöksenä oli
- 65. jäniksen ja lehtokurpan väkivaltainen kuolema. Hän helisti kelloa ja toimitti lasit jälleen täysiksi. Helle eneni. Herrat tempoivat tyytymättöminä tuolejaan ja huusivat tarjoilijaa. Se haihtuu aivan heti, lohdutteli tämä. Ei täällä minusta ole lainkaan liian lämmintä, sanoi Fink levollisesti. Minun puolestani saatte vaikka lisätä uuniin. Mutta helle kävi aivan sietämättömäksi, herrat vimmastuivat ja huusivat Feronia itseään paikalle. Anton vastusti kiivaasti akkunan avaamista, koska kaikki olivat vielä hiostuneet tanssista; Fink selitti lämmön olevan aivan parahultaisen ja piti takin päällään. Anton joutui epätoivoon. Vihdoin hän turvautui viimeiseen keinoon, riisui itse takin hartioiltaan saadakseen ystävän seuraamaan esimerkkiä. Heti Fink tekikin samoin, laski takkinsa huolellisesti tuolinselälle ja katsahti hymyillen Antoniin, joka hyvin jännitettynä oli vaarinottanut hänen liikkeitään. Ei se kirja takin taskussa ole, sanoi Fink nyökäten hänelle. Hukka vaiva, keksi jotain parempaa. Anton avasi ikkunan. En enää yritäkään, hän vastasi alakuloisesti; sinua en kuitenkaan kavaluudessa voita. Kestä sinä vain loppuun asti, sanoi Fink. Zernitz kertoili somia kaskuja; Tönnchen tiesi valheellisia juttuja tanssijattarista; pikku Lanzau joi itsensä humalaan. Vihdoin Fink koputti pöytään. Kuulkaapa nyt tarkoin. Tahdoin pitää tämän asian salassa, mutta se on mahdotonta, vääryys huutaa taivaaseen asti.
- 66. Anton riuhtaisihe pystyyn. Fritz, pidä suusi kiinni! Ole hiljaa, mokoma uuninlämmittäjä! huusi Fink. Kuulkaahan, hyvät herrat, minä olen tänään löytänyt ruskeiden neitosten salaisen päiväkirjan ja olen sitä selaillut perusteellisesti. Hurraa, antakaa kuulua! huusivat kaikki herrat. Siinä on varmastikin värssyjä, huusi Zernitz. Pelkkää hullutusta siinä taitaa olla! huusi Tönnchen. Alaikäisten letukkain haaveiluja ja ilkeyksiä. Anton oli raivoissaan. Hullutuksia siinä tosiaankin on, ja värssyt minusta ovat kehnoja. Kuulkaahan, Zernitz, mitä tekemistä teillä on ollut pikku Laran kanssa? Ei mitään, vastasi luutnantti oudostellen; olenhan tanssinut pari kertaa hänen kanssaan, siinä kaikki. Sekin riittänee, jatkoi Fink mietiskelevästi. Pikku Theone parka! Olen lukenut erään runon, jonka tuo pieni kreivitär on teille sepittänyt. No niin, ettehän tekään ole miehistä pahimpia, mutta enpä olisi otaksunut mahdolliseksi, että teistä voidaan niin ihaillen puhua. Näyttäkääpä minulle, pyysi Zernitz kiihkeästi. Täälläkö? kysyi Fink paheksuen, tämän hurjan ryövärijoukon läsnäollessa? Vaikka te ette erikoisesti suosikaan pikku Laraa, joka tänään hädässään näytti minusta ihmeen suloiselta, niin eihän teillä
- 67. kuitenkaan ole syytä saattaa tyttöparan puhdasta haaveilua täällä joka miehen hammasteltavaksi. Oikeassa olette, sanoi Zernitz. Mutta kahdenkesken ollessanne te näytätte sen minulle. Varmasti, vastasi Fink. Tiedättehän etten minä sääli noita otuksia, joilla hameet tuskin ulottuvat polven alapuolelle, ja jos mikä maailmassa jättää minut kylmäksi, niin ovat ne tuollaiset puolikasvuiset haihattelevat tyttöletukat. Mutta annettakoonpa totuudellekin arvonsa; nuo tytöt, jotka ovat yhdessä pitäneet tätä päiväkirjaa, ovat kerrassaan kelpo tyttöjä, eikä kirjassa ole lainkaan ilkeyksiä. Hän kääntyi serkku-Baldereckin puoleen: Teidän serkustanne puhutaan joka sivulla niin lemmekkäästi ja sydämellisesti, että se tuntuu yhtä liikuttavalta kuin ansaitultakin. — Ankarin arvostelu langetetaan minusta itsestäni; minua näet nimitetään viheriävarpuseksi. Tuohon tapaan jatkuen vihkonen taitaa käydä jokseenkin pitkäpiimäiseksi, huomautti Benno Tönnchen. Niinpä kyllä, sanoi Fink, jollei teitä huvita kuulla, mitä Hildegard Sait on teistä kirjoittanut. Paljon hyvää se ei taida sisältää, virkkoi Benno uteliaana. Eipä niinkään; hän puhuu teistä sävyssä, joka on omiaan todella surettamaan teidän tuttavianne. Teitä sanotaan suureksi ja hiljaiseksi sieluksi, kasvonne ovat muka miehekkään voiman mallikuva. Runoilijatar on keksinyt teidän olevan täynnä tietoja, henkevyyttä ja älykkäisyyttä; ja hän kysyy, eikö sellainen mies ole liian korkealla
- 68. tasolla alentuakseen huomaamaan hennon tyttösen taipumusta. Nytpä kysyn teiltä kaikilta, että miten ihmeellä voi niin viisas tyttö kuin Hildegard Sait siinä määrin erehtyä, että käy kaikessa hiljaisuudessa palvomaan teidän veroistanne miestä? Olettehan te tosin viimeistä pulloa tyhjennettäessä aika vekkuli, Benno, mutta jos minä olisin tyttö ja etsisin itselleni ihannetta, niin mieluummin valitsisin pähkinäpihdin epäjumalakseni kuin teidät. Tönnchen veti suunsa hymyyn. Onko siinä meistäkin jotain? kysyi herra von Werner, hänkin vihreitä, neljän ihanan sisaruksen veli, Rothsattelin lähinaapuri, tosin vastaleivottua aatelia, mutta hyvin rikas ja kasvanut keskellä molempien perheiden kateutta ja riitaisuuksia. Teistä on sangen vähän, Fink vastasi, vain kaksi riviä. Hän otti kirjan esiin ja selaili sitä etsivästi. — Anton puristeli käsiään nyrkkiin pöydän alla. — Kas tässä! Surkea taivaan sallima, Lenore rakastaa ja koettaa turhaan verhota sydäntään. Ja lemmitty kuuluu vihollisleiriin. Oo, Georg W. Sitten seuraa piste ja kolme huutomerkkiä. Fink piilotti kirjan jälleen. Anton rauhoittui. Mitään tuollaista ei kirja voinut sisältää; myöskin hän näki Finkin sierainten värähtelevän, mikä oli pettämätön merkki siitä, että hän hautoi koiranjuonia. Zernitz työnsi lasinsa syrjään ja huudahti: Minusta on epähienoa, että me tällaisessa paikassa pidämme pilanamme, mitä tytöt ovat sydämensä sisimmässä tunteneet. Minä olen samaa mieltä, huusi Benno Tönnchen kiihkeästi. Minä myös, toisti Werner.
- 69. Teidän täytyy sinetöidä kirja ja lähettää se takaisin omistajalleen, lausui lehtisammakko. Oi te tuntehikkaat poropeukalot, huudahti Fink naljaillen, teidän sydämenne sulaa kohta voiksi, kun vain hienot sormet raapivat teidän harjaspäitänne. Tahtoisinpa nähdä minkälaista naamaa olisitte näyttäneet, jos olisin lukenut kirjasta jotain päinvastaista. — Niinpä niin, kukapa teistä Shakespearea tuntee! Kreivitär Lara ja Hildegard ovat liiaksi hienotunteisia kirjoittaakseen mitään sellaista, joita te ilkeydessänne olisitte kirjassa kernaasti nähnyt, huusi Zernitz. Rothsattelin neiti on tosin ylpeä, jatkoi Werner, mutta mitäpä syytä hänellä olisikaan sanoa minusta mitään muuta kuin täyttä totta. Olen aina pitänyt häntä kelpo tyttönä, joka hyvin ansaitsee tulla herran rehellisen nuoren miehen vaimoksi. Fink nyykäytti hänelle hyväksyvästi päätään, otti kirjan jälleen esiin ja katsahti ylös kattoon. Minkävuoksi minua ei siirretä tältä syntiseltä maankamaralta parempien olentojen joukkoon? Minähän olen seraafi, mutta kukaan ei sitä älyä eikä tule koskaan uskomaankaan, kaikkein vähimmin hameväki. Kas tässä, Anton, ota huostaasi tämä kirja! Ei sitä ole valloitettu uuninlöylyn avulla, eikä viekoittelemalla eikä väkivalloin; vaan tanssikurssin herrain vapaaehtoisesta päätöksestä se palautetaan lukematta omistajilleen. Anton sieppasi joutuin kirjan, riensi Feronin konttorihuoneeseen, kirjoitti paperilapulle: Fink on lukenut joitakin sivuja, mutta hän pitää suunsa kiinni; kukaan muu ei ole vilkaissutkaan sen sisään; sitten hän kääri kirjan ja kirjelapun paperiin, sinetöi sen ja lähetti
- 70. käärön erään Feronin miehen myötä iltamyöhällä kreivitär Laran asuntoon, taottuaan moneen kertaan lähetin päähän, että tämän oli kaikin mokomin tunkeuduttava yövartijan ja portinvartijan käsien läpi kallisarvoisine kääröineen aina makuuhuoneen kynnykselle asti, sillä hän otaksui hyvällä syyllä Theone paran valvovan tuskaisena ja kastelevan kyynelillään mustat kiharansa märäksi tappurakuontaloksi. Juominki jatkui sitten entistä menoaan. Mutta höyryävän kuuma huone, väkevä juoma ja eräänlainen mietiskelevä mieliala useimmilla herroista teki istunnosta lopun varhemmin kuin Fink oli odottanut. Vihdoin hänkin nousi pöydästä, herätti nukahtaneen viinurin ja sanoi Antonille: Maksa lasku. Lähtiessään sitten Antonin kanssa patikoimaan kotia hän sanoi: Ole levollinen, Tony, kaikkihan tietysti oli silkkaa valhetta, mitä kirjasta oli lukevinani. Itse asiassa oli siihen kasattuna niin paljon ilkeyttä kuin parvi turturikyyhkysiä vain kykenee keksimään. Kyllähän sen huomasinkin, vastasi Anton mielissään, ja saatpa nähdä, että ensi tanssitunnilla nuo tuttavasi tulevat vimmatusti liehittelemään kaunokaisiaan. Jokunen heistä tulee lopulta vielä naimaankin sen lemmityn, jonka tänä iltana hänelle annoin. Taitaakin olla parasta, että oikein todenteolla antaudun parittajan ammattiin. Anton vaikeni mieli myrtyneenä. Älä sure suotta, jatkoi Fink hyvätuulisesti; tuletpa itsekin antamaan siunauksesi noille naimiskaupoille. Mitä sinä muuten pidit äskeisistä herroista? Katsoppas, sanoi Anton, mitä he yleensä puhelevat, se tuntuu minusta usein aivan arkipäiväiseltä, mutta heillä on itseluottamusta
- 71. ja varmaa ryhtiä, jota he eivät menetä yltiöpäisiksi tullessaankaan. No niin, sanoi Fink, onhan tuossakin perää; mutta muuten he ovat joutilaassa juoksussaan kaiken maailman naisserkkujen kanssa ja ratsastusvimmassaan surkastuneet peräti. Kokonaisuutena se ihmisluokka, johon he kuuluvat, on otettava varoittavaksi esimerkiksi siitä millainen ihminen ei saa olla pyrkiessään huvitteleimaan ja huvittamaan toisia. Heidän hulluttelunsa ei ole lystikästä, ja heidän lystikkäisyytensä on kurjaa; parissa vuodessa he ennättävät käydä aivan noloiksi ja sietämättömiksi kuin hapan ja huonosti käynyt viinimehu. Tuo Tönnchen alkaa jo maistua aika tunkkeutuneelta. Minua haluttaa suuresti näyttää heidät ensi kerralla sinulle tuiki humaltuneina. Älä puhu noin rivosti, pyysi Anton. Ah, sinä poika parka, sanoi Fink. Suljehan ulko-ovi perästämme ja anna minulle kukkaroni takaisin. Sinä olet tänään yksinäsi maksanut ison laskun, sanoi Anton. Pyydän ettet olisi niin avokourainen, sillä sinä vain nöyryytät toisia. Ole levollinen, Anton, vastasi Fink; he huvittavat minua, siksipä on oikein ja kohtuullista, että minä maksan heidän puolestaan. Toivon ettet koskaan tule maksamaan minun puolestani, virkkoi Anton. En tulekaan, sanoi Fink. Sinulla pitää oleman etuoikeus olla oma rahainhoitajasi; minä tyydyn siihen, että kannat meidän yhteistä portinavaintamme ja että käyt polttamaan sikarisi loppuun minua luonani, sillä aikaa kuin minä riisuudun. — Mitähän kello on?
- 72. Lyö kohta kaksi, vastasi Anton paheksuvasti. Sitten olemme varmastikin viimeiset yökulkijat. Kun tulin tähän vanhaan talorähjään, ei se tuntunut jaksavan sulattaa moisia hurjasteluja. Kun ensi kertaa aamuyöstä työnsin tämän jättiläisavaimen lukkoon, pelkäsin noiden vanhain muurien sortuvan minun syntiseen niskaani. Nyt ne ovat jo siihen tottuneet, samoin koira, piharengit ja talon isäntä. Usein jään vain senvuoksi myöhään kaupungille, jotta voisin pyörähdyttää mokomata poroporvarillista järjestystä edes vähäsen sijoiltaan. * * * * * Kun Hildegard Sait kyynelkostean yövalvonnan jälkeen vasta aamupuoleen rupesi suunnittelemaan maatamenoa, havahdutti hänet jälleen Theone Laralta tuotu kirje, jonka alkuosassa Theone mustan korpisulan välityksellä lausui mielipiteenään, että tässä matoisessa maailmassa ei ollut enää tilaa hänelle, ja jälkiosassahan perusteli tämän mielipiteensä kutsumalla Hildegardin ja Lenoren seuraavana iltapäivänä suklaatikekkereihin juhlimaan yhdessä päiväkirjan onnellista pelastumista. Tässä ruskeitten sotaneuvottelussa pohdittiin innokkaasti, oliko kirja ja missä määrin tullut häväistyksi sen kautta, että julkeat miessilmät olivat kurkistaneet sen sivuille. Hirveätä oli, että Fink oli lukenut siitä kappaleita. Mutta myöskin Wohlfartilla oli se ollut käsissä, ja sangen luultavaa oli, että hän oli lukenut sen kannesta kanteen. Lenore oli puolestaan vallan varma siitä, että Wohlfart ei ollut edes kurkistanut kansien sisäpuolelle. Mutta Hildegard väitti, että hänkin oli vain mies, ja ettei parahinkaan mies olisi kyennyt vastustamaan moista kiusausta. Pitkän pohtimisen perästä päätettiin
- 73. hänet panna koetteelle. Jos hän on katsahtanut kirjan sisään, sanoi Lenore, niin ensiksi hän tietysti on nähnyt nimilehden. Nimilehteä hän olisi vaaratta saanut katsella, huomautti eräs ruskeista lintusista. Minä olin kieltänyt häntä edes avaamasta kirjaa, lausui Lenore, ja minä tiedän, että hän on totellut minua. Te saatte kaikki kuulla, mitä hän tulee vastaamaan kysymyksiini. Kun Anton saapui seuraavalle tanssitunnille, kävi Lenore puolueensa etunenässä häntä vastaan; hänen ilmeensä oli hyvin huolestunut, ja kaikki ruskeat riiputtivat paitansa ja koettivat näyttää murheellisilta. Ah, herra Wohlfart, mitä te olettekaan tehnyt! Kirja, jonka lähetitte Theonelle, ei ollutkaan hänen päiväkirjansa, vaan jonkin herran muistikirja. Onko se mahdollista? huudahti Anton peljästyneenä. Jo aivan ensi sivulla oli viime kuun 29 päivänä päivätty lasku uudesta hännystakista, ja seuraavan päivän kohdalla merkintä pullosta punaviiniä ja uudesta kannusparista. Siitä kirjasta ei meille ollut hiukkaakaan apua, päivittelivät ruskeat neitoset päätään pudistellen ja katsellen alakuloisesti lattiaan. Anton yritti puolustautua. Fink otti tuon punakantisen kirjan liivintaskustaan ja antoi sen käteeni, ja minä lähetin sen heti sinetöitynä matkaan. Sitten on herra von Fink vaihtanut kirjat, jatkoi Lenore ripitystään. Mutta miksi ette katsahtanut sen sisään, hän lisäsi moittivasti, edes nimilehteen?
- 74. Eihän minulla ollut lupa, Anton huudahti. Olinhan luvannut teille, etten avaisi ollenkaan kirjaa. Minä huudan Finkin tänne. Seis, odottakaahan hetkinen, sanoi Lenore. Onko hän siis katsonut kirjaan vai eikö? hän kysyi riemuitsevasti parveltaan. Ihmettelevä ja ihaileva Ei! kajahti, kaikkien huulilta. Jääkää vain tänne, herra Wohlfart, oikean kirjan te meille palautitte. Jotkut meistä epäilivät, voisiko yksikään mies — edes tekään — jättää käsistään tyttöjen kirjoittamaa päiväkirjaa siihen edes vilkaisematta; minä sanoin teidän pystyvän siihen, ja nyt olen sen todistanut ystävilleni. Kiitän teitä hyvästä luottamuksestanne, huudahti Anton ilahtuneena. Teistä uskon kaikkea, mikä on hyvää ja rehellistä, vastasi Lenore ja katsoi sydämellisen luottavaisesti nuorakaista silmiin. Sinä iltana vallitsi tanssiseurassa oikein juhlatunnelma. Aina kotiljongin alkuun saakka Antonia ympäröi kukkea piiri nuoria neitosia, jotka kohtelivat häntä liikuttavan tuttavallisesti, ja kun tuli aika, jolloin naiset jakoivat värillisiä nauharuusuja herroille, koristettiin Antonin hännystakin rinnukset aina liepeeseen saakka, jotta hän näytti ainakin yhtä korealta kuin mannermaan kaikkein kirjavatakkisin hovimarsalkka. Mutta vielä suurempiakin ihmeitä tapahtui. Vihreä puolue uhkasi kannattajien puutteessa luhistua tykkänään kokoon. Zernitz, Werner ja pikku Lanzau tanssivat tänä iltana yksinomaan ruskeitten kanssa. Hildegard Saltin täytyi elää hirvittävä puolituntinen Pähkinäpihdin parissa, joka kohteli häntä tosi ritarillisesti, voipa sanoa, lämpimän
- 75. tuntehikkaasti, kasaten tyttö poloisen pään päälle sen kautta tulisia hiiliä; Lenore puolestaan sai kärsiä kohteliaita rynnäkköjä lehtisammakon, Georg Wernerin ja pikku Lanzaun taholta, jotka kaikki kolme olivat yht'äkkiä tulleet siihen vakaumukseen, ettei Lenore ei ollut aivan arvoton vastaanottamaan heidän ritarillista kunnioitustaan. Yksinpä Eugeniekin osotti tänään ruskeille vilpitöntä sydämellisyyttä, hän riippui pitkän aikaa Lenoren kainalossa ja suuteli Theonea tämän poislähtiessä ylen liikutettuna molemmille poskille. Ja rouva von Werner istahti paroonitar Rothsattelin viereen, lupasi käydä huomenissa vieraisilla kaikkien tyttäriensä kanssa, pyysi lupaa ottaa Georginsakin mukaan ja puheli lakkaamatta siitä, kuinka onnelliset hänen lapsensa tulisivat olemaan ensi kesänä, kun tanssikurssi oli saattanut heidät niin läheiseen ystävyyteen Lenoren kanssa. Sanalla sanoen, tanssi-illan yleisleima oli kerrassaan muuttunut. Lukuunottamatta vihreitä neitosia, jotka pahoittelivat kavaljeeriensä uskottomuutta, vallitsi salissa herttainen, ihmisrakkautta uhkuva mieliala, jonka keskipisteinä olivat ruskean puolueen jäsenet. Asemansa muuttuminen saattoi heidät aivan hämilleen; Baldereckin herttaisuus, entisten vihollisherrain huomaavaisuus tuntui kyllä mieluiselta; mutta, ah! onneaan he eivät kuitenkaan kyenneet täysin siemauksin nauttimaan, povessaan he tunsivat tunnonvaivojen tuhannet neulanpistokset, nähdessään kaikkitietävän Finkin hirvittävän hahmon leijailevan heidän ympärillään laajassa kehässä. Yhdellä sanalla hän voi repiä rikki heitä kietovan lumouksen. — Koko illan Fink pysytteli kaukana kaikista päiväkirjan kirjoittajista; vasta aivan lopulla hän astui Lenoren luo ja sanoi tälle kiusoitellen: Eikö neiti Eugenie ole tänään oikein herttainen? Myönnän kyllä, että hän on tunteeton, mutta ehkäpä tuo pieni vika vuosien mittaan muuttuu aivan päinvastaiseksi hyveeksi.
- 76. Lenore katsahti häneen hämillään. Tulkaa kanssani Theone Laran luo, sanoi hän vihdoin. Sinne tultuahan huudahti liittolaistensa kuullen: Meidän kaikkien tulee kiittää herra von Finkiä, ja me pyydämme häntä vast'edeskin vaikenemaan päiväkirjan suhteen, niinkuin hän on tähänkin asti tehnyt. Lupaan sen mielelläni, vastasi Fink, mutta yhdellä ehdolla. Joku uhri täytyy minun nielaista. Minun täytyy saada tietää, kuka naisista on kyhännyt säkeet erään viiniköynnöstä esittävän kuvan alle. Jokuhan minulla täytyy olla, jota saan vihata kaikesta voimastani, josta saan joka tilaisuudessa puhua kaikkea mahdollista pahaa, joka saa nahallaan maksaa teidän kevytmielisyytenne, kun kaikesta älykkäisyydestänne huolimatta annoitte tuon kallisarvoisen todistuskappaleen joutua minun käsiini. Mainitkaa minulle joku nimi, ja minä puolestani lupaan teille vapaaehtoisesti, etten koskaan enää lausu riviäkään kirjan sisällyksestä kellekään vieraalle. Ruskeitten ryhmässä syntyi hätäytynyt liike; jokainen pelkäsi joutuvansa tuon kostonhimoisen intiaanin raadeltavaksi. Lenore katsahti Hildegardiin, joka oli pelosta palttinanvalkeana, ja sanoi sitten innokkaasti: Minä piirustin kuvan ja sanelin säkeet, jotka ystäväni kirjoitti sen alle. Koskapa olette nähnyt koko kauhistuksen, niin pyydän teiltä anteeksi. Enemmän en voi tehdä; ja jos nyt aiotte kostaa siitä minulle, niin on minun koetettava parhaani mukaan kestää vihaanne. Hyvä, sanoi Fink hymyillen, minä käynkin kostamaan. Tästä illasta alkaen olen vihaava teitä sydämeni pohjasta. Muuten on minun mieluista todeta, että kaikkein haihtuvaisinkin inhimillisistä tunteista, tyttöjen välinen ystävyys, voi innoitta sen uhriksi sortuneet onnettomat todella sankarillisiin tekoihin. — Ah, neiti Hildegard, eikö
- 77. Benno Tönnchen teistäkin ole oikein kiltti pienokainen? Eikä hänen vartalonsakaan ole hullumpi. Ehkä liian täyteläinen sanonette, mutta juuri tuo täyteläisyys tekee hänet hänen rotunsa niin puoleensavetäväksi. Viimeisenä seurauksena tästä odottamattoman onnellisesta illasta eli, että ruskea puolue päätti palkita Wohlfartin uskollista ritaripalvelusta jollain erinomaisella tavalla. Pitkän pohtimisen perästä tultiin yksimielisiksi siitä, että Theone yhdessä ystävättäriensä kanssa virkkaisi hänelle komean rahakukkaron. Jo seuraavana aamuna ostettiin sitä varten silkkilankaa ja helmiä. Päästäkseen mukaan työhön päätti Lenorekin vartavasten opetella virkkaamisen taidon. Ja jopa loistelikin kukkaron toinen puolisko valmiina ruskean- ja kullanhohtoisena, kun sattui tapauksia, jotka odottamatta ehkäisivät työn päättämisen. 3. Surullinen kokemus on, että taivahan vallat eivät kauan salli ihmislapsen huolettomasti nauttia korkean tunnelman tuottamata onnea. Ne ovat järjestäneet asiat niin ovelasti, että melkein aina jokin jänne laukeaa sielussamme, kohta kun ne ovat virittäneet vieressä olevan jänteen mahtavasti soinnahtelemaan. Luonnollisena seurauksena tästä on värähtelevä epäsointu. Sellaisen tylyn kohtelun esineeksi joutui Antoninkin sielu… Ensimmäisenä epäsoinnun oireena oli, että koko konttori edelleenkin seurasi hyvin vaanivin katsein Antonin elämässä tapahtunutta muutosta. Takapihan puolella oltiin siitä yksimieliset, että siitä lähtien kuin sankarimme oli ruvennut käymään tanssitunneilla, oli hänen olemuksensa huomattavasti muuttunut eikä
- 78. lainkaan edukseen. Tosin hän vapaahetkinään seurusteli virkaveljiensä kanssa vähemmän kuin ennen, vietti monet illat poissa kotoa, ja kun hän kerran oli mukana yhteisessä illanvietossa, saattoi hän olla jonkun verran hajamielinen ja kenties hiukan pisteliäskin toisten herrain hänelle tuttujen pikku heikkouksien johdosta. Hänen terve järkensä kyllä varjeli häntä rupeamasta mahtailemaan hienossa seurapiirissä äkkiä saavuttamistaan voitoista ja ikävystyttämästä toisia niiden kertomisella; mutta hän ei kuitenkaan voinut välttyä tekemästä vertailuja kotipiirissä vallitsevan sävyn — jonka hän hyvin tunsi — ja ylhäisen rouvan salongissa vallitsevan sävyn välillä, joka hänelle itselleen oli ihkasen uutta. Työtoverit pitivät hänen entistä suurempaa vaiteliaisuuttaan ylpeytenä, hänen tiheää poissaoloaan kevytmielisenä hurjastelemisena; ja siten joutui Anton, joka ennen oli ollut koko talon suosikki, juuri tämän seikan vuoksi sitä ankaramman arvostelun esineeksi. Itse puolestaan hän tunsi maltillisempani toverien pidättyväisyyden ja jyrkempien kouraantuntuvan kylmyyden hyvin kipeästi ja valitti itsekseen joutuneensa lemmettömän kohtelun alaiseksi. Siitäpä johtui, että hän nekin illat, jolloin hänellä ei ollut aihetta poistua kotoa, vietti miltei yksinomaan Finkin parissa, ja että nämä molemmat piankin muodostivat pienen ylimyksellisen nurkkakunnan, joka oli vihamielisessä suhteessa toisiin herroihin.
- 79. Antonia tällainen suhde masensi enemmän kuin hän tahtoi itselleenkään myöntää; hän kärsi siitä työpöytänsä ääressä, omassa huoneessaan, jopa päivällispöydässäkin kadunvartisessa talossa. Harvemmin kuin ennen hänen työtoverinsa häntä puhuttelivat; tarvitessaan jotain tietoa Jordan ei enää kääntynyt hänen puoleensa vaan Baumannin; kun rahastonhoitaja aamiaishetken lyötyä astui etukonttoriin, ei hän enää pysähtynyt Antonin pöydän ääreen; ja kun Specht kesken kauppakirjeittensä kirjoittamista tapansa mukaan käännähti äkkiä rajusti ympäri tuolillaan ja teki jonkin merkillisiä kysymyksiään, niin hän kohdisti sen tosin entistä ahkerammin Antonille, mutta tämän asemaa ei suinkaan parantanut, kun Specht muka kuiskaten karjui hänen korvaansa: Onko totta, että herra von Blogilla on papurikkovaljaikko? tahi: Täytyykö rouva von Baldereckin luo mennessä vetää jalkaansa kiiltonahkasaappaat vai kelpaako kengät? Kaikista tylyimmin Antonia kohteli hänen vanha suosijansa Pix. Liika suuri suvaitsevaisuus ei ollut koskaan ollut tämän tarmokkaan herran vahvoja puolia, ja jostakin hämärästä syystä hän piti nykyistä Antonia konttorin, ison vaa'an ja soolowhistin petturina. Hänen tapanaan oli viettää syntymäpäiväänsä mahdollisimman juhlallisessa muodossa. Silloin hän kutsui kaikki uskotut ystävänsä, joista Anton oli ensimmäisiä, illaksi huoneeseensa, jossa hän päivän kunniaksi tarjosi viiniä ja sokerikakkua; jälkimmäinen oli vartavasten tilattu sokerileipurilta ja esiytyi vuosi vuodelta yhä suuremmassa koossa. Näinä viikkoina sattui jälleen hänen syntymäpäivänsä, ja vaikka Pix olikin viime aikoina ollut merkillisen vaitelias Antonia kohtaan, oli tämä kuitenkin varustautunut viettämään illan hänen luonaan ja oli sen vuoksi vastannut kieltävästi herra von Zernitzin kutsuun. Varhain ennen konttorituntien alkua hän meni Pixin huoneeseen onnittelemaan tätä. Herra Pix kuunteli onnitteluja hyvin kylmäkiskoisesti eikä kutsunut
- 80. Antonia illaksi luokseen. Pöydästä noustua Anton kohtasi tuon suunnattoman sokerikakun, joka leipurin oppipojan hartioilla työläästi kapusi takapihan portaita ylöspäin; ja konttorissa hän tajusi herra Spechtin puheista, että tällä kertaa kaikki toverit olivat kutsutut pyhittämään sen päivän muistoa, jolloin herra Pix maailmaantulollaan oli täyttänyt tuntuvan aukon luomakunnassa. Niin — kaikki muut olivat kutsutut, paitsi hän ja Fink. Syystäkin Anton käsitti tämän syrjäyttämisen tahalliseksi loukkaukseksi. Mutta hän antoi sen vaikuttaa itseensä kipeämmin kuin olisi ollut tarpeellistakaan. Ja päällepäätteeksi piti vielä Spechtin suihkata hänelle kuin minäkin hauskana salaisuutena, että Pix oli sanonut, että sellainen nuori herra, joka seurusteli luutnanttien kanssa ja illoin irstaili Feronissa, ei ollut sopiva seuralainen vakavien kauppamiesten piiriin. Kun hän tänä iltana istui yksin huoneessaan ja kuuli toverien iloista hälinää Pixin puolelta, valtasi hänet ahdistava ja tuskallinen tunne, jota ei kyennyt hajoittamaan mikään niistä loistavista kuvista, jotka viime aikana olivat joutohetkinä täyttäneet hänen mielensä, ei edes ihanin ja rakkainkaan niistä. Itsekään hän ei ollut lainkaan tyytyväinen itseensä, vaan koetti oikein kasata kiduttavia syytöksiä itseänsä vastaan. Hän oli tosiaankin muuttunut. Eihän hän tosin työssä ollessaan päästänyt itseään vallan laukeamaankaan, mutta työstään hänellä oli varsin vähän iloa, pikemminkin se tuntui taakalta. Olipa sattunut, että hän liikekirjeissä oli unohtanut jonkin tärkeän asian mainitsematta, olipa hän pari kertaa kirjoittanut vääriä hintamääriäkin, jolloin Jordan oli kuivasti huomauttaen jättänyt kirjeet hänelle takaisin. Hänelle johtui myöskin mieleen, ettei kauppias ollut viime aikana ollenkaan välittänyt hänestä ja hänen toimistaan, ja että Sabine oli joku päivä sitten portailta tavatessaan tervehtinyt häntä kylmäkiskoisemmin
- 81. kuin tavallisesti. Ja aivan äskettäin, kun täti oli valittanut yörauhansa häiriytyneen sen kautta, että joku oli niin myöhään ja kolistellen tullut portista sisään, olivat kaikki toverit katsoneet häneen moittivasti. Yksinpä uskollinen Karlkin oli tuonaan kysynyt — Antonin mielestä sangen ivallisesti — oliko hänellä portinavain mukana tanssitunneille lähtiessään. Tällaisten tukalien muistelmien vaivaamana Anton kävi kirjoituspöytänsä ääreen ja rupesi silmäilemään pientä yksityistä kassakirjaansa. Viime viikkoina hän ei ollut lainkaan merkinnyt siihen menojaan; senvuoksi hän tarttui kynään ja alkoi hätäytyneenä laskea niitä yhteen sormistaan laskuista ja muististaan. Hirmustuen hän huomasi velkainsa nousevan summaan, jota hän ei mitenkään kyennyt suorittamaan, jollei käynyt käsiksi vanhempainsa jättämään pikku perintöön. Hän tunsi itsensä perin, onnettomaksi. Tähän asti oli kohtalo soitellut hänelle vain hienoimpia säveleitään, nyt rupesivat toiset kielet pahoin särähtelemään. Ja epäsoinnun piti käydä vieläkin pahemmaksi. * * * * * Samana iltana palasi kauppias huonolla tuulella klubistaan kotia, vastasi kuivasti Sabinen tervehdykseen ja rupesi pitkin askelin mittelemään huoneen lattiaa. Mikä sinua vaivaa, Traugott? kysyi sisar. Veli astui hänen tuolinsa viereen. Tahdotko tietää, millä tapaa Fink on saattanut suojattinsa rouva von Baldereckin taloon? Sinähän olet niin iloinen noiden molempien ystävyydestä. Nyt hän on kutonut kokoon oikean valheiden verkon ja tehnyt kokemattomasta Wohlfartista julkean seikkailijan. Hän kertoisitte erään vanhemman
- 82. Welcome to our website – the perfect destination for book lovers and knowledge seekers. We believe that every book holds a new world, offering opportunities for learning, discovery, and personal growth. That’s why we are dedicated to bringing you a diverse collection of books, ranging from classic literature and specialized publications to self-development guides and children's books. More than just a book-buying platform, we strive to be a bridge connecting you with timeless cultural and intellectual values. With an elegant, user-friendly interface and a smart search system, you can quickly find the books that best suit your interests. Additionally, our special promotions and home delivery services help you save time and fully enjoy the joy of reading. Join us on a journey of knowledge exploration, passion nurturing, and personal growth every day! ebookbell.com