126,99 €
Mehr erfahren.
- Herausgeber: John Wiley & Sons
- Kategorie: Wissenschaft und neue Technologien
- Sprache: Englisch
The Internet of Edges is a new paradigm whose objective is to keep data and processing close to the user. This book presents three different levels of Edge networking: MEC (Multi-access Edge Computing), Fog and Far Edge (sometimes called Mist or Skin). It also reviews participatory networks, in which user equipment provides the resources for the Edge network. Edge networks can be disconnected from the core Internet, and the interconnection of autonomous edge networks can then form the Internet of Edges. This book analyzes the characteristics of Edge networks in detail, showing their capacity to replace the imposing Clouds of core networks due to their superior server response time, data security and energy saving.
Sie lesen das E-Book in den Legimi-Apps auf:
Seitenzahl: 339
Veröffentlichungsjahr: 2022
Ähnliche
Table of Contents
Cover
Title Page
Copyright Page
Introduction
I.1. The Edge
I.2. The digitization of companies
I.3. The different levels of the Edge
I.4. Conclusion
I.5. References
1 Edge Architectures
1.1. The three levels of Edge Networking
1.2. Edge Computing architectures
1.3. Security and domain name system on Edge
1.4. The digital infrastructure of the participatory Internet
1.5. Conclusion
1.6. References
2 MEC Networks
2.1. The MEC level of 5G architecture
2.2. 5G
2.3. 5G Edge
2.4. Conclusion
2.5. References
3 Fog Networks
3.1. Fog architectures
3.2. Fog controllers
3.3. Fog and the Internet of Things
3.4. Wi-Fi in the Fog’s digital infrastructure
3.5. The new generation Wi-Fi
3.6. The next generation of mobile Wi-Fi
3.7. Private 5G for Fog Networking
3.8. Conclusion
3.9. References
4 Skin Networks
4.1. The architecture of Skin networks
4.2. Virtual access points
4.3. Participatory Internet networks
4.4. Conclusion
4.5. References
5 Ad hoc and Mesh Networks
5.1. Ad hoc networks
5.2. Routing
5.3. Mesh networks
5.4. Participatory networks
5.5. Local services
5.6. The digital infrastructure of the Internet of the Edges
5.7. Conclusion
5.8. References
6 Applications of the Internet of Edges
6.1. Civil security and defense applications
6.2. Applications of the Internet of Things
6.3. The tactile Internet
6.4. Telecom applications
6.5. Industry 4.0
6.6. The smart city
6.7. Conclusion
6.8. References
7 Vehicular Networks
7.1. Communication techniques for vehicular networks
7.2. Vehicular Ad hoc NETworks
7.3. Connected and intelligent vehicles
7.4. The MEC and the VEC
7.5. Intelligent transport systems (ITS)-G5
7.6. 5G V2X
7.7. The VLC
7.8. Conclusion
7.9. References
8 Virtualization of the Internet of Edges
8.1. Network virtualization
8.2. Virtualization on the Edge
8.3. Using virtual networks on the Edge
8.4. Mobile Edge Computing
8.5. Conclusion
8.6. References
9 Security
9.1. Cloud of security on the Edge
9.2. Secure element
9.3. Blockchain
9.4. Conclusion
9.5. References
10 The Example of Green Communications
10.1. The Green PI solution
10.2. The Edge Cloud
10.3. The IoE
10.4. The IoE platform
10.5. Use cases: IoT in constrained environments
10.6. IoT in motion
10.7. Massive IoT
10.8. The advantages
10.9. References
11 Deployment of the Participatory Internet
11.1. The deployment
11.2. The Green Cloud
11.3. Scaling up
11.4. Energy savings
11.5. Security
11.6. Wi-Fi and LTE hybridization
11.7. Conclusion
11.8. References
12 The Future
12.1. The short-term future
12.2. The medium-term future
12.3. The long-term future
12.4. Participatory Internet and IPV6
12.5. References
List of Authors
Index
End User License Agreement
List of Tables
Chapter 3
Table 3.1.
The main standards of the IEEE802.11 working group
Chapter 5
Table 5.1.
Ad hoc protocols
Chapter 7
Table 7.1.
Priority classes
Table 7.2.
Default values for EDCA parameters
Table 7.3.
Baud rate, modulation scheme and code rate in 802.11p
Chapter 10
Table 10.1.
Benefits of the IoE platform
List of Illustrations
Introduction
Figure I.1.
An Internet of Edges.
Figure I.2.
The digitization of companies.
Figure I.3.
The IT environment of an enterprise
Figure I.4.
The different levels of Edge Computing and the Cloud Computing.
Chapter 1
Figure 1.1.
The distributed cloud environment.
Figure 1.2A.
The basic elements of OIF (a) and CNCF (b)
Figure 1.2B.
The basic elements of the EECC.
Figure 1.3.
The evolution of architectures.
Figure 1.4.
Some modules of the OpenStack open-source software.
Figure 1.5.
The main properties of the CNCF architecture.
Figure 1.6.
The global architecture of the CNCF.
Figure 1.7.
The Kubernetes orchestrator.
Figure 1.8.
CaaS and FaaS architectures.
Figure 1.9.
Service architectures.
Figure 1.10.
An eSIM environment
Figure 1.11.
Participatory Internet infrastructure (©Green Communications).
...
Chapter 2
Figure 2.1.
The 5G environment.
Figure 2.2.
The different 5G accesses.
Figure 2.3.
Transition from 4G to 5G.
Figure 2.4.
Key technology components of 5G
Figure 2.5.
Coexistence of different waveforms associated with F-OFDM.
Figure 2.6.
A massive MIMO system.
Figure 2.7.
The long-term 5G core network.
Figure 2.8.
The basic slicing of the 5G core network (©3GPP).
Figure 2.9.
The fully centralized Cloud-RAN architecture.
Figure 2.10.
The partially distributed Cloud-RAN architecture.
Figure 2.11.
The MEC environment as defined by ETSI
Figure 2.12.
The eNodeB or gNodeB of a mobile network antenna.
Figure 2.13.
The eNodeB offset in a MEC data center.
Chapter 3
Figure 3.1.
The object-related architecture broken down into two levels.
Figure 3.2.
Skin and Fog servers.
Figure 3.3.
A Fog data center to manage a hospital.
Figure 3.4.
Industry 4.0 processes.
Figure 3.5.
An example of hardware from the Open Compute Project.
Figure 3.6.
Case study for the location of a controller.
Figure 3.7.
The main functions of a Fog controller.
Figure 3.8.
Proximity marketing.
Figure 3.9.
The networks of the Internet of Things.
Figure 3.10.
Architecture of a Wi-Fi network
Figure 3.11.
Frequency used for Wi-Fi
Figure 3.12.
The SDMA.
Figure 3.13.
The main characteristics of Wi-Fi 6 (IEEE 802.11ax).
Figure 3.14.
The main characteristics of Wi-Fi 7 (IEEE 802.11be).
Figure 3.15.
The new features of Wi-Fi 7.
Figure 3.16.
The arrival of WiGig technologies.
Figure 3.17.
The aggregation of two carriers, one licensed and the other Wi-Fi
...
Figure 3.18.
Disaggregation of the base model to O-RAN.
Figure 3.19.
The O-RAN architecture.
Figure 3.20.
3GPP architecture for 5G and Wi-Fi convergence (©3GPP).
Figure 3.21.
The simultaneous use of 5G and Wi-Fi.
Chapter 4
Figure 4.1.
A Skin data center
Figure 4.2.
A virtualized HNB.
Figure 4.3.
Context of a network around a Skin data center.
Figure 4.4.
A Skin data center environment for creating a participatory Intern
...
Figure 4.5.
A participatory Internet network
Figure 4.6.
Mesh networks and ad hoc networks
Chapter 5
Figure 5.1.
Ad hoc network
Figure 5.2.
Coverage extension by an ad hoc network
Figure 5.3.
Direct communication between machines in an ad hoc network
Figure 5.4.
Routing through intermediate nodes
Figure 5.5.
Structure of the Hello package
Figure 5.6.
Structure of the TC package
Figure 5.7.
A participatory network.
Figure 5.8.
The participatory network in Figure 5.7 after the disappearance of
...
Figure 5.9.
Some examples of connections between devices.
Figure 5.10.
Example of link bitrates.
Chapter 6
Figure 6.1.
Example of the interconnection of three clusters.
Figure 6.2.
Some 5G-specific applications.
Figure 6.3.
Architecture of the tactile Internet.
Figure 6.4.
Extensions of the participatory Internet.
Figure 6.5.
An Internet of Edges infrastructure
Figure 6.6.
Generations of industrial manufacturing.
Figure 6.7.
Solutions for moving to Industry 4.0.
Figure 6.8.
The smart city infrastructure.
Chapter 7
Figure 7.1.
A vehicular network.
Figure 7.2.
Communication between vehicles: V2V.
Figure 7.3.
V2V in the context of 5G.
Figure 7.4.
Communication by light between two vehicles.
Figure 7.5.
An example of a VANET network.
Figure 7.6.
An intelligent vehicle
Figure 7.7.
Vehicles in cohorts.
Figure 7.8.
The MEC.
Figure 7.9.
The MEC and the VEC.
Figure 7.10.
An example of frames and subframes of a C-V2X communication. Comp
...
Chapter 8
Figure 8.1.
Three virtual routers.
Figure 8.2.
Network virtualization.
Figure 8.3.
Virtual machines in the Edge’s digital infrastructure.
Figure 8.4.
OpenFlow signaling.
Figure 8.5.
The OpenFlow frame.
Figure 8.6.
A physical OpenFlow controller.
Figure 8.7.
Network device virtualization.
Figure 8.8.
Example of geolocation managed by the MEC data center.
Figure 8.9.
Example of an augmented reality application on the MEC.
Figure 8.10.
Example of video analytics.
Figure 8.11.
Example of content optimization.
Figure 8.12.
Example of a data cache system and DNS resolution.
Figure 8.13.
Example of performance optimization.
Figure 8.14.
Different placements of the MEC data center.
Figure 8.15.
Architecture of a MEC server.
Chapter 9
Figure 9.1.
A Cloud of security.
Figure 9.2.
A firewall for the 2020s.
Figure 9.3.
Some topologies for firewalls on the Edge.
Figure 9.4.
Big Data Analytics-based security environments.
Figure 9.5.
A secure element server in a Cloud of security (©EtherTrust).
...
Figure 9.6.
EAP smart card authentication procedure.
Figure 9.7.
Location of the profiles on an eSIM card.
Figure 9.8.
Evolution of SIM card size.
Figure 9.9.
The different security solutions.
Figure 9.10.
Relationships between TSM participants.
Figure 9.11.
The solution without TSM.
Figure 9.12.
Securing with a local secure element.
Figure 9.13.
Securing with secure external elements.
Figure 9.14.
Securing a Cloud of secure elements.
Figure 9.15.
Advantages of the eSIM solution.
Figure 9.16.
Security architecture by secure external elements.
Figure 9.17.
Securing virtual machines.
Figure 9.18.
Securing an electronic payment.
Figure 9.19.
Functioning of a blockchain
Chapter 10
Figure 10.1.
Edge Cloud’s IoT Platform.
Figure 10.2.
Internet of Edges platform.
Figure 10.3.
Architecture of the Green PI platform (green color for what we pr
...
Figure 10.4.
IoE in maritime pollution monitoring.
Figure 10.5.
IoE as a heterogeneous swarm of drones, robots, vehicles and peop
...
Figure 10.6.
IoE in smart hospitals and telemedicine.
Chapter 11
Figure 11.1.
YOI router (©Green Communications).
Figure 11.2.
A deployed participatory Internet (©Green Communications).
Figure 11.3.
Green Communications’ Edge Cloud: the Green Cloud (©Green Communi
...
Figure 11.4.
The My Network application (©Green Communications).
Figure 11.5.
The Chat application.
Figure 11.6.
Green Communications IoE.
Figure 11.7.
Example of a network topology
Figure 11.8.
State diagram showing the possible states and transitions for Sta
...
Figure 11.9.
SaS performance results.
Figure 11.10.
Hybridization technique of Green Communications.
Figure 11.11.
Example of use.
Chapter 12
Figure 12.1.
A distributed SDN environment.
Figure 12.2.
Medium-term solution on the Edge.
Figure 12.3.
Horizontal and direct communication versus operator network commu
...
Guide
Cover Page
Title Page
Copyright Page
Introduction
Table of Contents
Begin Reading
List of Authors
Index
WILEY END USER LICENSE AGREEMENT
Pages
iii
iv
ix
x
xi
xii
xiii
xiv
xv
xvi
xvii
xviii
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
235
237
238
240
241
242
243
SCIENCES
Networks and Communications, Field Director – Guy Pujolle
Cloud Networking, Subject Head – Kamel Haddadou
Edge Networking
Internet of Edges
Khaldoun Al AghaPauline LoygueGuy Pujolle
First published 2022 in Great Britain and the United States by ISTE Ltd and John Wiley & Sons, Inc.
Apart from any fair dealing for the purposes of research or private study, or criticism or review, as permitted under the Copyright, Designs and Patents Act 1988, this publication may only be reproduced, stored or transmitted, in any form or by any means, with the prior permission in writing of the publishers, or in the case of reprographic reproduction in accordance with the terms and licenses issued by the CLA. Enquiries concerning reproduction outside these terms should be sent to the publishers at the undermentioned address:
ISTE Ltd
John Wiley & Sons, Inc.
27-37 St George’s Road
111 River Street
London SW19 4EU
Hoboken, NJ 07030
UK
USA
www.iste.co.uk
www.wiley.com
© ISTE Ltd 2022The rights of Khaldoun Al Agha, Pauline Loygue and Guy Pujolle to be identified as the authors of this work have been asserted by them in accordance with the Copyright, Designs and Patents Act 1988.
Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s), contributor(s) or editor(s) and do not necessarily reflect the views of ISTE Group.
Library of Congress Control Number: 2022941477
British Library Cataloguing-in-Publication DataA CIP record for this book is available from the British LibraryISBN 978-1-78945-068-2
ERC code:PE7 Systems and Communication EngineeringPE7_8 Networks (communication networks, sensor networks, networks of robots, etc.)SH3 The Social World, Diversity, PopulationSH3_12 Communication and information, networks, media
Introduction
I.1. The Edge
The Internet of Edges, the subtitle of this book, is the formulation in simple terms of a new generation of networks: an infrastructure of communications and services that is realized at the edge of the network. The Internet of Edges is an interconnection of Edge networks. There are three levels of Edge networks: Skin networks, Fog networks and Mobile/Multi-access Edge Computing (MEC) networks. Skin networks connect clients to each other on the far edge, and the network nodes that support services are within a few meters of the users. Fog networks are more for company use. In a sense, they are replacing local area networks. However, their structure is a little different since the objective is to connect the company’s employees to a small data center located within the company. MEC networks originate from the Edge networks of telecommunications operators. These are the networks using 5G and connecting customers to an operator’s data center located very close to a 5G antenna. We have represented an Internet of Edges in Figure I.1.
Skin, Fog and MEC networks can themselves be participatory networks or not. A participatory network is built by interconnecting the electronic machines of its participants, which can be a user’s terminal, a nearby box, a drone, a robot or a vehicle, etc. Participatory networks are distinguished from other types of networks by their totally distributed aspect. A non-participatory network is a network where services are centralized in a cloud, usually far away, to which customers must connect to obtain a service. In participatory networks, all services are distributed, whether they are infrastructure services or application services. The distance between the user and the data center hosting the service is comparable to the range of Wi-Fi communication. The machines that participate in the network bring their resources and host services that will be available to all users of the network.
Figure I.1.An Internet of Edges.
The participatory Internet is a participatory network using a fully distributed Transmission Control Protocol/Internet Protocol (TCP/IP) environment. It is a new paradigm that brings together autonomous direct communications between mobile nodes and the concept of Edge Computing. The user of the participatory Internet is no longer a consumer of the environment’s resources but, on the contrary, participates in the realization and life of the digital infrastructure of this new Internet concept. The network is alive; it forms and deforms itself according to its participants. It replaces the classic Internet while using the same protocols in a distributed way and more or less powerful Edge data centers. IP compatibility allows it to connect or disconnect from the Internet without altering its operation. Of course, disconnection can stop some services that require centralized Internet servers. On the other hand, the participatory Internet allows mobility, independence, autonomy, instant deployment and strong security. The participatory Internet is perfectly applicable to companies to manage their information system and their applications while bringing increased security to their IT environment. The participatory Internet is very well suited to telecom operators by allowing immediate extensions of their network. This new concept is also very interesting for integrators by facilitating the introduction of mobile digital infrastructure to customers who need it. The architecture related to the participatory Internet is designed to realize a new world of mobility. It adapts to vehicular networks and is expected to become the standard for the mobile Internet that is bound to emerge with connected vehicles. Many forecasts show that drivers will switch from driving to connecting to the Internet as soon as the vehicle becomes autonomous. As a result, the amount of data flowing through the Internet from mobile devices could reach almost the same value as that from fixed customers. The participatory Internet is also suitable for robot networks, drone networks and all networks in which humans, machines, or objects are in motion. It is particularly suitable for tactical cells and smart spaces. This new paradigm of the participatory Internet is associated with ad hoc and mesh technologies in hybrid mode and with data and application processing at the Edge. It is an integral part of the digitization of enterprises and will be integrated into Edge environments that have two other components: the MEC and Fog. These two other components are characterized by more powerful and totally fixed data centers.
The new Internet of Edges generation with participatory networks can be referred to as the uberization of telecommunications since it allows the creation of a network using only machines belonging to users, whether they are the general public or companies. However, this architecture based on participatory networks can also be implemented by telecommunications operators through their users with their smartphones or their Internet box. An Internet of Edges network based on participatory networks can be created using machines that integrate the TCP/IP protocol stack and hosting servers to provide services to clients connected to this network. The machines can be large servers, small computers, or even smartphones. Any electronic device with a processor, memory and network interface is suitable to participate in the creation of a participative network. This network can be interconnected to other networks with gateways to globalize it. The interconnection can be made with other participatory networks but also with more classical Internet networks. In the latter case, a user can go and look for centralized services on the Internet.
Currently, the global Internet works well and satisfies many needs, but it also consumes many resources because of the distance of users from the data centers that have the functions and data needed to provide the services requested. To send a message to its neighbor, the message must travel long distances to reach the server and return to the sender’s neighborhood. This also leads to the creation of huge and energy-consuming data centers that are difficult to control and secure. The idea of deglobalizing the Internet makes it possible to use nearby resources to provide a service, like Airbnb or Uber, where anyone with a resource (house, car, etc.) can offer it to enrich the accommodation or transportation offer.
In the rest of this book, we will take an in-depth look at the Internet of Edges but also the reasons for its introduction to easily achieve the digitization of businesses and the industrial world. We will also examine the other two components of Edge Computing, the MEC and the Fog, and their cooperation with the level closest to the user, the Skin level, with a particular interest in participatory networks.
I.2. The digitization of companies
The digitalization of business and industry is the main reason for the extension of digital infrastructures, especially at the edge, whether it be through MECs, Fog data centers, or the participatory Internet, which is located at the level closest to the user.
We have represented in Figure I.2 the different elements of this digitization from networks to applications.
At the digital network level, we find the three main components based on virtualization: Software-defined networking/Network functions virtualization (SDN/NFV), 5G and Cloud-Native technologies. The participatory Internet is also part of digital networks and can use virtualization techniques and be associated with 5G. It can also use the solutions provided by Cloud Native.
Figure I.2.The digitization of companies.
Digital operations are increasingly in demand for automation to enable cost reduction by having autonomous networks that can automatically handle the configuration, control and management of the digital infrastructure. Open Source represents the second way to lower the costs of infrastructure. It is necessary to move towards open application interfaces so that network and application modules can be easily interconnected, with the possibility of marketing more efficient proprietary modules that could replace the open-source modules. At the top level of the digital experience, there are real-time applications and generic applications that enable easy implementation of services, optimizations and reliability. Finally, at the highest level, the digital services that can be expected from the entire infrastructure such as high-definition video, applications that bring the virtual such as augmented reality, intelligent applications for the city, building or home, autonomous vehicles, tactile services such as remote surgery and finally the Internet of Industrial Things to realize applications of Industry 4.0.
Overall, three levels are defined in the new IT environments: the digital infrastructure, the platforms and, at the highest level, the services. Figure I.3 describes the IT environment of companies and the industrial world.
Figure I.3.The IT environment of an enterprise
At the level of digital infrastructure, there are two sub-levels: the technical infrastructure composed of hardware, mainly data centers of various sizes, antennas and cables, whether fiber optic or wire cables. The second sub-layer corresponds to the digital software infrastructure, which gathers all the functions of the infrastructure, such as routing, switching, firewall, signal processing, etc.
The layer above forms the platform on which the applications run. This platform is made up of software that enables the simple development of services. Finally, the highest layer, the services layer, contains all the virtual machines that will provide services to the users of the digital infrastructure.
I.3. The different levels of the Edge
Edge Networking is a subset of Edge Computing that comprises the data centers located at the network’s edge. Edge Networking deals with the digital infrastructure consisting of infrastructure and application services. These services are integrated into data centers that form a distributed multi-cloud. Three categories of data centers make up this distributed multi-Cloud: the MEC, the Fog and the Skin, as shown in Figure I.4.
Figure I.4.The different levels of Edge Computing and the Cloud Computing.
The MEC is the telecom operators’ version of virtualizing all the equipment between the customer and the data center: the processes related to the antenna, but also the intermediate equipment such as the Internet boxes. The Cloud-Radio Access Network (Cloud-RAN), and its evolution into the Software-Defined-RAN (SD-RAN) for dynamic management of the sharing of physical layer functions, is part of this MEC environment of Edge Networking.
The Fog represents the data centers of companies that are positioned on their premises and that allow the virtualization of the Local Area Network (LAN) and Wide Area Network (WAN) access equipment and, of course, all the company’s business processes. It can be virtualized on a MEC data center if the company outsources its IT and networks to an operator. The Fog allows the realization of virtual Customer Premises Equipment (vCPE) and the support of SD-WAN.
The Skin is a new vision, watched very closely by some of the big Web companies such as Google, which sees an opportunity not to leave room for operators and companies to develop by themselves on the Edge. Indeed, this solution would allow the introduction of proximity marketing which is the future of advertising. The Skin allows tracking of a customer wherever they are, and with a very good knowledge of their instantaneous environment, it is easy to send them customized advertising. Each customer receives a different advertisement than other users. Proximity marketing represents a completely personalized solution to advertising.
These three tiers of Edge networks will co-exist in a distributed multi-cloud. In addition, some companies and organizations will trust their carrier to manage networks and applications from the MEC, others will want to keep control of their IT and network and others will want to be able to set up mobile smart spaces like those found in civil security, events, or mobile worksites.
I.4. Conclusion
The two concepts of centralization and distribution alternate over time, moving from one to the other approximately every 10 years. Centralization arrived in the early 2010s with Cloud Networking and SDN to enable simplified controls and intelligence that was only known to be done centrally. The 2020s should see a return to distribution with an overlay of the two solutions between 2020 and 2025.
This distribution should come back with the paradigm of participatory networks and the Internet of Edges based on participatory networks. The Internet of Edges is a concept that takes up the basic properties of the Internet as they were defined at its birth with a distribution of powers to allow a strong resilience and good security by avoiding points of high sensitivity and distributed management.
In the rest of this book, we will examine the current state of Edge networks coming from the 2010s thinking and reaching maturity and the transition to next-generation distributed networks that will be dominant from 2025.
I.5. References
Aazam, M., Harras, K.A., Zeadally, S. (2019). Fog computing for 5G tactile industrial Internet of Things: QoE-aware resource allocation model.
IEEE Trans. Ind. Inform
., 15, 3085–3092.
Ahmed, A. and Ahmed, E. (2016). A survey on mobile edge computing.
10th International Conference on Intelligent Systems and Control (ISCO)
, Coimbatore, India, 7–8 January 2016.
Ahmed, E., Gani, A., Khan, M.K., Buyya, R., Khan, S.U. (2015). Seamless application execution in mobile cloud computing: Motivation, taxonomy, and open challenges.
Journal of Network and Computer Applications
, 52, 154–172.
Ahvar, E., Orgerie, A., Lébre, A. (2019). Estimating energy consumption of cloud, fog and edge computing infrastructures.
IEEE Transactions on Sustainable Computing
, 1–12. doi: 10.1109/TSUSC.2019.2905900.
Anttalainen, T. (1999).
Introduction to Telecommunications Network Engineering
. Artech House, Norwood.
Byrne, P. (2017).
Computer Networking
. Larsen and Keller Education, New York.
Clark, M.P. (1997).
Networks and Telecommunications: Design and Operation
. Wiley, New York.
Comer, D.E. (2013).
Internetworking with TCP/IP Volume 1: Principles, Protocols, and Architecture
, 5th edition. Prentice-Hall, Hoboken.
Czarnecki, C. and Dietze, C. (2017).
Reference Architecture for the Telecommunications Industry: Transformation of Strategy, Organization, Processes, Data, and Applications
. Springer, Berlin.
Davies, D.W. and Barber, D.L.A. (1973).
A Communication Networks for Computers
. Wiley, New York.
Elkhodr, M. and Hassan, Q. (2017).
Networks of the Future: Architectures, Technologies, and Implementations
. Chapman & Hall, London.
Fajjari, I., Aitsaadi, N., Pióro, M., Pujolle, G. (2014). A new virtual network static embedding strategy within the Cloud’s private backbone network.
Computer Networks
, 62(7), 69–88.
Frahim, J. and Josyula, V. (2017).
Intercloud: Solving Interoperability and Communication in a Cloud of Clouds
. Cisco Press, Indianapolis.
Freeman, R. (1999).
Fundamentals of Telecommunications
. Wiley, New York.
Goralski, W. (2017).
The Illustrated Network, 2nd Edition: How TCP/IP Works in a Modern Network
. Morgan Kaufmann, Burlington.
Hanako, A. (2016).
Communications and Computer Networks
. Clanrye International, New York.
Hérold, J.F. and Guillotin, O. (2015).
Informatique industrielle et réseaux
. Dunod, Malakoff.
Kurose, J.F. and Ross, K.W. (2016).
Computer Networking
. Addison-Wesley, New York.
Lohier, S. and Présent, D. (2016).
Réseaux et transmissions – Protocoles, infrastructures et services
. Dunod, Malakoff.
Macchi, C. and Guilbert, J.-F. (1988).
Téléinformatique
. Dunod, Malakoff.
Mckenzie, V.D. (2017).
Mobile Networks: Concepts, Applications and Performance Analysis
. Nova Science Publishers Inc., Hauppauge, New York.
Musa, S. and Wu, Z. (2017).
Aeronautical Telecommunications Network: Advances, Challenges, and Modeling
. CRC Press, Boca Raton.
Oliver, N. and Oliver, V. (2006).
Computer Networks: Principles, Technologies and Protocols for Network Design
. Wiley, New York.
Peterson, L. and Davie, B.S. (2017).
Computer Networks, 4th Edition: A Systems Approach
. Morgan Kaufmann, Burlington.
Pujolle, G. (2019).
Les réseaux
. Eyrolles, Paris.
Robertazzi, T.G. (2017).
Introduction to Computer Networking
. Springer, Berlin.
Robinson, D. (2017).
Content Delivery Networks: Fundamentals, Design, and Evolution
. Wiley, New York.
Rojas-Cessa, R. (2017).
Interconnections for Computer Communications and Packet Networks
. CRC Press, Boca Raton.
Severance, C.R. (2015).
Introduction to Networking: How the Internet Works
. Create Space Independent Publishing Platform, Scotts Valley.
Tanenbaum, S. (2010).
Computer Networks
. Prentice-Hall, Hoboken.
Valdar, A. (2017).
Understanding Telecommunications Networks
. IET Press, London.
White, C. (2015).
Data Communications and Computer Networks: A Business User’s Approach
. Cengage, Boston.
1Edge Architectures
1.1. The three levels of Edge Networking
The four-tier distributed multi-cloud architecture is becoming the standard. The levels correspond to four types of data centers that form the basis for the digitalization of various companies and organizations. Large data centers symbolize the Cloud with almost infinite computing and storage power. However, these data centers are far from the users with high latency to reach them. Many applications cannot run with these latencies. In addition, the energy consumption per client is very high since the information flows have to pass through many intermediate machines. However, these data centers are indispensable for applications that require power and for particularly consistent storage. The most important data centers belong to the GAFAM companies (Google, Apple, Facebook, Amazon and Microsoft), but also the BATX (Baidu, Alibaba, Tencent and Xiaomi).
For several years, data centers at the edge of networks have become increasingly important for many reasons. Latency allows the introduction of new services that are becoming preponderant for the digitalization, intelligence and security of companies, administrations and public services. For example, Industry 4.0, with IIoT (Industrial IoT), smart grid control, short-latency applications, etc. are examples of these services. These applications have been highlighted by a number of companies. These applications have been highlighted by 5G, which must have an important place in the digital infrastructure of the Edge. Next-generation Wi-Fi should not be forgotten as convergence with 5G has now clearly started. The diagram in Figure 1.1 shows the Edge environment of a network.
Figure 1.1.The distributed cloud environment.
Three levels of Edge Computing are being implemented to define a distributed multi-Cloud. The most advanced in its definition and deployment concerns the MEC (Mobile/Multi-access Edge Computing) data centers of 5G operators, which enable the virtualization of numerous telecom devices and the introduction of new services that are well described in the 3GPP specifications and which concern the Internet of Things, mission-critical services with high reliability and very high speeds in mobility.
The second level, defined a long time ago, concerns Fog Networking. The word Fog was introduced by Cisco in 2013. Initially, Fog Computing was used to manage streams from objects to pre-process them before sending them to a central Cloud. Today, Fog Networking is linked to enterprise data centers that virtualize local network equipment such as routers, switches, firewalls, authentication servers and many middleboxes with control and management functions. The vCPE (virtual customer premises equipment) uses Fog data centers.
The third level is the one that is the focus of this book. It has various names that have not yet stabilized, such as Very Edge Networking, the home data center, Mist-Computing or Skin Computing, which we will use in the following. These structures are generally related to the participatory Internet, although we can see star systems around a data center. For a participatory network, the tiny data centers are located very close to the user or even on the user. These are small but powerful servers that support virtualization. These Skin data centers must be light, autonomous and consume very little energy since they can be mobile and follow the users like firemen during a fire. The advantages come from the extremely short latency time, their low energy consumption and their mobility, not to mention their security by disconnecting them from the Internet thanks to their autonomy as soon as necessary.
The four levels we have described overlap, and no one level is expected to be dominant in the future. The distributed Cloud was born from this observation. To optimize performance, services must be positioned as virtual machines (VM) or containers that users need at the right level of the architecture. This level is highly dependent on the application and its performance, control, management and security characteristics.
These digital infrastructure services, whether infrastructure or application services, are divided into microservices encapsulated in virtual machines or containers. We will see in the next chapter the platform architectures and this decomposition into microservices. An even finer decomposition is emerging in the early 2020s with the decomposition into functions. A microservice is, therefore, a set of functions.
1.2. Edge Computing architectures
The three main architectures associated with the services, carried by the data centers of the different levels of the Edge, come from the OIF (Open Infrastructure Forum), the Cloud Native of the CNCF (Cloud Native Computing Foundation) and the EECC (European Edge Computing Consortium). In Figures 1.2A and B, we have depicted the main elements of these three environments.
Figure 1.2A.The basic elements of OIF (a) and CNCF (b)
Figure 1.2B.The basic elements of the EECC.
Notably, the Telco Edge Cloud (TEC) is an initiative that started in 2020 and aims to define an Edge architecture allowing interoperability between operators.
An intelligent digital infrastructure on the Edge requires the deployment of a distributed multi-Cloud distributed framework using 5G and Wi-Fi integration. This generation also relies on infrastructure and application services that must make businesses and organizations more agile, flexible and independent of the web giants. Figure 1.3 shows the evolution of these architectures up to the 2020s that we are interested in here.
In Figure 1.3, looking at the architecture of the 2020s, we see many changes. The physical infrastructure, or technical infrastructure, now contains only three elements: the antennas, optical fibers and data centers. Everything else is virtualized in the data centers. All the boxes have disappeared from this architecture. Indeed, the functions associated with these boxes are virtualized in one of the data centers with urbanization to determine its level Edge or Cloud. This vision of the physical infrastructure is not relevant for participatory networks at the endpoints, which can be mobile and which we will describe a little further on.
Figure 1.3.The evolution of architectures.
Above the physical infrastructure is the digital infrastructure, which contains all the functions related to the physical infrastructure to make it work. These are the functions found in boxes such as Wi-Fi boxes, routers, switches, firewalls, etc. Again, these functions are positioned at the different levels associated with Skin, Fog, MEC or Cloud data centers.
Infrastructure services correspond to all the basic functions that can be used by the application services. Infrastructure services are the services provided to meet general requirements, including technical and software solutions, such as interoperability, security, middleware and network or distributed systems services. Infrastructure services also include the management and control of application services. These services are becoming part of the application services as they integrate more and more functions that used to be put directly into the application service but are now available directly to all applications that want to use them rather than being found in parallel at the level of each application.
We are now going to take a closer look at the two most widely used digital infrastructures, which initially came from the major Web industries but now have the support of almost all industries: the OIF and the CNCF.
OIF took over the OpenStack Alliance in 2020. Open Stack is a Cloud management system that is widely used by Cloud providers and operators. However, the Cloud is not enough to define an infrastructure. This is the reason for the birth of OIF: to add to the Cloud and to virtualization the ingredients to build a complete infrastructure.
OpenStack was born in 2010 with the merger of two projects, Rackspace (Storage) and NASA (Compute). OpenStack is developed in Python and distributed under the Apache 2.0 license. The versions started with A, then B, etc. The latest are:
– Queens (2018, 02);
– Rocky (2018, 08);
– Stein (2019, 04);
– Train (2019, 10);
– Ussuri (2020, 05);
– Victoria (2020, 10);
– Wallaby (2021, 04).
OpenStack contains a large number of modules that allow you to build a complete Cloud. Some of these modules are described in Figure 1.4. They include:
– heat: orchestration component;
– neutron: network and addressing management;
– nova: management of the computing resources;
– cinder: block storage;
– swift: object storage;
– glance: disk image service;
– ceilometer: metric-based telemetry;
– keystone identity service.
OpenStack is a complete environment that is increasingly being introduced in commercial software. However, while it supports virtualization techniques well, it needs to be complemented by other open-source Cloud infrastructure software. For example, IOF integrates Magma for Edge infrastructure, offering both cellular (5G) and Wi-Fi services and the OpenInfra Labs. It includes, for example, the Wenjua project that provides services to shorten the introduction time needed to integrate artificial intelligence. The OIF architecture includes, of course, Kubernetes to orchestrate the containers.
Figure 1.4.Some modules of the OpenStack open-source software.
The second major architecture is the Cloud Native Computing Foundation (CNCF). This is a project of the Linux Foundation that was founded in 2015 to help advance container technology and align the technology industry around its evolution. Founding members include Google, CoreOS, Mesosphere, Red Hat, Twitter, Huawei, Intel, Cisco, IBM, Docker, Univa and VMware. Today, the CNCF is supported by nearly 500 members.
CNCF technology projects are cataloged with a Sandbox, Incubated and Graduated maturity level in ascending order. The CNCF process incorporates projects as Incubated projects and then aims to progress them to Graduated, which involves a level of process and technology maturity.
