ETSI GS AFI 002 V1.1.1 (2013-04)

Autonomic network engineering for the self-managing Future Internet (AFI); Generic Autonomic Network Architecture (An Architectural Reference Model for Autonomic Networking, Cognitive Networking and Self-Management)

DGS/AFI-0002

General Information

• Status: Published
• Publication Date: 11-Apr-2013
• Technical Committee: AFI - Autonomic network engineering for the self-managing Future Internet
• Current Stage: 12 - Completion
• Due Date: 17-Apr-2013
• Completion Date: 12-Apr-2013

Standards Content (Sample)

Group Specification
Autonomic network engineering for the self-managing
Future Internet (AFI);
Generic Autonomic Network Architecture
(An Architectural Reference Model for Autonomic Networking,
Cognitive Networking and Self-Management)
Disclaimer
This document has been produced and approved by the Autonomic network engineering for the self-managing Future Internet
(AFI) ETSI Industry Specification Group (ISG) and represents the views of those members who participated in this ISG.
It does not necessarily represent the views of the entire ETSI membership.

2 ETSI GS AFI 002 V1.1.1 (2013-04)

Reference
DGS/AFI-0002
Keywords
architecture, autonomic networking, cognition,
generic, model, network, self-management

ETSI
650 Route des Lucioles
F-06921 Sophia Antipolis Cedex - FRANCE

Tel.: [phone]  Fax: [phone]

Siret N° 348 623 562 00017 - NAF 742 C
Association à but non lucratif enregistrée à la
Sous-Préfecture de Grasse (06) N° 7803/88

Important notice
Individual copies of the present document can be downloaded from:
http://www.etsi.org
The present document may be made available in more than one electronic version or in print. In any case of existing or
perceived difference in contents between such versions, the reference version is the Portable Document Format (PDF).
In case of dispute, the reference shall be the printing on ETSI printers of the PDF version kept on a specific network drive
within ETSI Secretariat.
Users of the present document should be aware that the document may be subject to revision or change of status.
Information on the current status of this and other ETSI documents is available at
http://portal.etsi.org/tb/status/status.asp
If you find errors in the present document, please send your comment to one of the following services:
http://portal.etsi.org/chaircor/ETSI_support.asp
Copyright Notification
No part may be reproduced except as authorized by written permission.
The copyright and the foregoing restriction extend to reproduction in all media.

© European Telecommunications Standards Institute 2013.
All rights reserved.
TM TM TM
DECT , PLUGTESTS , UMTS and the ETSI logo are Trade Marks of ETSI registered for the benefit of its Members.
TM TM
3GPP and LTE are Trade Marks of ETSI registered for the benefit of its Members and
of the 3GPP Organizational Partners.
GSM® and the GSM logo are Trade Marks registered and owned by the GSM Association.
ETSI
3 ETSI GS AFI 002 V1.1.1 (2013-04)
Contents
Intellectual Property Rights . 7
Foreword . 7
1 Scope . 8
1.1 Reading Guide and Snapshots of the Reference Model . 10
1.2 Snapshots of the Reference Model (a nutshell view) . 11
2 References . 14
2.1 Normative references . 15
2.2 Informative references . 15
3 Definitions and abbreviations . 19
3.1 Definitions . 19
3.2 Abbreviations . 25
4 Definition of the AFI Architectural Reference Model of a Generic Autonomic Network
Architecture and its fundamental requirements . 29
4.1 Definition . 29
4.2 The Generic Autonomic Network Architecture (GANA) Reference Model . 29
5 Properties of a Generic Autonomic Network Architecture Reference Model . 31
6 Enabling concepts and mechanisms . 32
6.1 Information and knowledge management mechanisms . 32
6.2 Cognitive mechanisms . 32
6.3 Service Models and Service Discovery Mechanisms . 32
6.4 Network Governance Mechanisms . 33
6.5 Capability discovery mechanisms . 33
6.6 Embodiment Mechanisms . 33
6.7 System Modelling. 33
6.8 Modularity/Composability . 34
6.9 Platform-Independent Execution/Capability Model of a network element . 34
6.10 Cooperation Mechanisms . 34
6.11 (Cross-Layer) Monitoring Methods and Techniques . 34
6.12 Fault-Detection, Fault-Diagnosis/Localization, Fault-Removal/Repair and Recovery Techniques and
Mechanisms . 35
6.13 Bootstrapping Mechanisms . 35
6.14 Programmability . 35
7 Generic Autonomic Networking Architecture Reference Model . 36
7.1 Principles & Motivations: Why it is required . 36
7.2 A review of today's best known approaches to Autonomic Networking . 37
7.3 Why a holistic dimension for Architectural Reference Model . 38
7.3.1 GANA properties . 39
7.3.2 GANA Meta-Model . 39
7.3.3 GANA principles relating to structural aspects . 40
8 Core Concepts of the GANA Reference Model . . 40
8.1 Approach taken to developing the concepts . 40
8.1.1 Implementation Use Case as an illustration . 41
8.2 GANA Reference Model: Structure, Core concepts and Principles . 41
8.2.1 The four GANA basic abstraction levels and their associated types of Control Loops . 44
8.2.1.1 GANA Level-1: the Protocol-Level . 45
8.2.1.2 GANA Level-2: the Function-Level . 46
8.2.1.3 GANA Level-3: the Node-Level . 47
8.2.1.4 GANA Level-4: the network's overall functionality . 48
8.2.2 GANA Functional Planes . 50
8.2.2.1 The Decision Plane . 50
8.2.2.2 Dissemination Plane . 51
ETSI
4 ETSI GS AFI 002 V1.1.1 (2013-04)
8.2.2.3 Discovery Plane . 52
8.2.2.4 Data Plane . 52
9 Architectural Principles for specificating Autonomic Behaviours (ABs) of Decision-Elements . 53
9.1 Definition of an Autonomic Behaviour (AB) . 53
9.1.1 High-Level State Transitions of a Decision-Element (DE) . 53
9.2 GANA Control Loops . 54
9.3 Protocol-intrinsic Control Loop and its associated Decision Element (DE) . 55
9.4 Example of Decision Element (DE) . 55
9.5 Structure of a GANA node and related DEs Hierarchy . 56
9.6 The Internal Interface/Reference Point within a GANA Node . 60
9.6.1 Specification (High level guidance) . 60
9.6.2 Implementation (high level guidance) . 60
9.7 Cross-Layering in GANA . 61
9.8 Combining Centralized and Distributed Decision-Making based Management in GANA . 62
9.8.1 Interfaces specification . 62
9.8.2 Illustration of interaction between fast control-loops in an NE and the slower outer loops in the

Knowledge Plane, using "routing function" case . 64
9.9 Reference Points (Logical Interfaces) within Network-Level-DEs interactions . 66
9.10 What needs to be standardized in the Autonomic Behaviours (ABs'): implementation guide . 67
9.10.1 Elements of a Control Loop . 68
9.10.2 Grouping the Autonomic Behaviour Specifications . 69
9.10.3 Specification aspects for Control-Loops . 69
9.11 Models of a Decision Element and a Managed Entity (ME) . 69
9.11.1 The Model of a Decision Element (DE), Models of Managed Entities (MEs) at GANA's lowest . 69
9.11.2 A Managed Entity (ME) at GANA's lowest layer (Variant -A) . 71
9.11.3 A Managed Entity (ME) that is an "evolved Protocol" or a Future Protocol Model in GANA-at
GANA's lowest layer (Variant - B) . 72
9.11.4 Enabling Programmability within MEs and DEs . 73
9.11.4.1 Programmability: what it is about . 73
9.11.5 Assignment of Managed Entities (MEs) and their Configurable and Controllable Parameters to
specific Decision Elements (DEs) in GANA ("ME-Param"-mapped to-"1-DE") . 75
9.11.5.1 Concept of "ownership" in GANA . 75
9.11.6 GANA Hierarchy - Mapping of Managed Entities (MEs) and their Configurable and Controllable
Parameters to specific DEs (ME-Param -to- 1-DE Mapping) . 80
9.12 GANA as a Unifying Model. 82
9.13 Cognition and Knowledge Plane as part of GANA Decision Plane . 85
9.13.1 Overview and Basic Definitions . 85
9.13.2 What is Cognitive Networking?. 85
9.13.2.1 Cognitive process . 86
9.13.2.2 Cognitive Architecture . 86
9.13.2.3 Relationship to existing Networking Planes . 87
9.13.3 Impacts on the GANA Model . 88
9.13.3.1 Relationship to GANA Model (functional aspect) . 88
9.13.3.2 Support for Decentralization . 88
9.13.3.3 Implementation in DE's . 89
9.13.4 The need for an Information/Knowledge Sharing Overlay Network . 89
9.13.4.1 The Subscription Mechanism . 92
9.13.4.2 Bootstrapping . 92
9.13.4.3 Inter-Domain Information/Knowledge Sharing through ONIX-to-ONIX Interface . 92
9.13.5 Knowledge Plane as part of the Decision Plane of the GANA Model . 92
9.13.5.1 Knowledge Plane definitions . 92
9.13.5.2 The Knowledge Plane according to GANA . 94
9.13.5.3 GANA Decision Plane . 94
9.13.6 Possible Approaches to Deriving Knowledge for the Knowledge Plane . 96
9.13.6.1 Knowledge Representation and Translation Component . 97
9.13.7 Other perspectives on Knowledge Building and use by Knowledge Plane . 98
9.14 Possible approach to designing the internal modules of a Decision-Element (DE) . 98
9.15 Virtualization in GANA . 99
9.15.1 Autonomic capabilities in IT-based resources virtualization . 99
9.15.1.1 Autonomicity and Virtualization within GANA . 100
9.15.2 Autonomic capabilities in Virtual Networks . 102
ETSI
5 ETSI GS AFI 002 V1.1.1 (2013-04)
9.15.3 Mapping GENI into the GANA Model. 103
9.15.4 GENI virtualization approach when applied in GANA: Autonomic Virtual Router case . 104
9.15.4.1 OpenFlow and GANA. 105
9.15.5 Formalism for virtual resource request and negotiation. 106
9.16 Stability in Autonomic Networks : Stability in GANA . 106
9.16.1 Stability Issues in Autonomic Networking . 107
9.16.2 Designing for Stability . 107
9.16.2.1 Stable Autonomic Behaviours Design through Game Theory - From Theory to Theory . 108
9.16.2.1.1 How to Treat Stability via Analytical Methods? - A Game Theoretic Approach . 108
9.16.2.1.2 How to address stability via Game Theory? . 109
9.16.2.1.3 Addressing Stability in an Architectural Level - From Theory to Practice . 109
9.16.2.1.4 Hierarchy of Control-Loops (DEs) . 109
9.16.2.1.5 Concept of "Ownership" . 110
9.16.2.1.6 Separation of "Operating Regions" . 110
9.16.2.1.7 Model-based Techniques . 110
9.16.3 Addressing Stability at Runtime . 111
9.16.3.1 Autonomic-aware Metrics to Infer and Self-assess Stability . 111
10 Network Governance . 112
10.1 GANA Network Governance Interface . 112
10.1.1 GANA Network Profiles . 120
10.1.1.1 Vertical Decomposition . 122
10.1.1.2 Horizontal Decomposition . 123
10.1.1.3 Relation between Vertical and Horizontal Decomposition . 124
10.1.2 GANA Network Profile Files and their Relationships . 125
10.1.3 GANA Network Profiles and Policy creation, distribution and modification . 128
10.1.3.1 Network-Level Policies . 128
10.1.3.2 Node- and Function-Level Policies . 129
10.1.3.3 Routing Policy Example . 129
10.2 Capabilities Self-Description and Self-Advertisement . 132
10.2.1 General Considerations . 132
10.2.2 Capability Description Files and their Relationships . 134
10.3 Decision Notification in GANA for the "Human in the Loop" towards building Trust and Confidence . 134
10.4 Autonomic Services Management in a GANA compliant network . 134
10.4.1 Service provision scheme and its evolution . 134
10.4.2 Service Management. 136
10.4.3 Service Management and Network Management Levels Cooperation . 136
11 Mapping and Impact of GANA on today's Management Paradigms . 138
11.1 GANA Decision Plane and today's Management Plane . 138
11.2 GANA and today's Control Plane. 138
11.3 GANA and today's Data Plane . 138
11.4 GANA Mappings to the TMN Logical Layered Architecture (LLA) . 139
11.5 GANA-DEs vs TMN Architecture, and impact on EMS and NMS levels . 140
11.6 The FCAPS Framework in the GANA architecture . 142
11.7 GANA and Network Management Systems (NMS's) . 142
11.7.1 Migration/Co-Existence scenarios that would still accommodate SNMP/XML/HTTP-based
management within GANA Network_Level_DEs-driven management . 144
11.7.2 Link between a DE and Management application that is HTTPand XML based . 145
11.8 GANA and OSS's . 146
11.8.1 Requirement framework for a Policy-Based Management . 146
11.9 Network self-management based on capabilities of the network as described to the overlying OSS
processes. 148
11.10 Network-intrinsic autonomic management (in-network management) via DE-to-DE interactions across
nodes/devices . 149
12 Federation in GANA . 152
12.1 Definition of federation . 152
12.2 Approach on federation . 152
13 Reference Points relevant in GANA compliant networks . 153
Annex A (informative): AFI Top-Down & Bottom-up Methodology . 159
ETSI
6 ETSI GS AFI 002 V1.1.1 (2013-04)
Annex B (informative): Authors & contributors . 160
Annex C (informative): Acknowledgements . 162
Annex D (informative): Other Useful Information relevant in Knowledge Derivation,
Representation and Presentation to the GANA Knowledge Plane . 163
Annex E (informative): Bibliography . 165
History . 167

ETSI
7 ETSI GS AFI 002 V1.1.1 (2013-04)
Intellectual Property Rights
IPRs essential or potentially essential to the present document may have been declared to ETSI. The information
pertaining to these essential IPRs, if any, is publicly available for ETSI members and non-members, and can be found
in ETSI SR 000 314: "Intellectual Property Rights (IPRs); Essential, or potentially Essential, IPRs notified to ETSI in
respect of ETSI standards", which is available from the ETSI Secretariat. Latest updates are available on the ETSI Web
server (http://webapp.etsi.org/IPR/home.asp).
Pursuant to the ETSI IPR Policy, no investigation, including IPR searches, has been carried out by ETSI. No guarantee
can be given as to the existence of other IPRs not referenced in ETSI SR 000 314 (or the updates on the ETSI Web
server) which are, or may be, or may become, essential to the present document.
Foreword
This Group Specification (GS) has been produced by ETSI Industry Specification (ISG) Autonomic network
engineering for the self-managing Future Internet (AFI).
NOTE 1: The present document contains some content that was published earlier in some scientific publications
and has been re-used or evolved by the original authors (copyright is acknowledged and citations
provided). Examples include concepts related to the earlier versions of the GANA Model that has been
evolved in the present document.
NOTE 2: Where there are some figures taken from the public domain and have been included for providing
information/illustrations to readers or have been modified to fit into the architectural framework,
Copyright Acknowledgments to the authors and publishers have been added apart from referencing/citing
the original publisher. Acknowledged Copright owners include: Taylor and Fransis Group, LLC,
Auerbach Publications, publishers cited for MAPE, GENI, FOCALE, 4D, Knowledge Plane for the
Internet, CONMan, Deriving Knowledge for the Knowledge Plane (a NIST Draft), Concepts developed in
EC funded EFIPSANS project and other EC funded projects listed at the end of the present document in
acknowledgements appendix.
ETSI
8 ETSI GS AFI 002 V1.1.1 (2013-04)
1 Scope
The main objective of the present document is to define, iteratively, a generic, conceptual architectural reference model
intended to serve as guideline for the design of the future generation networks exhibiting autonomic characteristics or
capabilities. The technical content of the present document consists of a set of fundamental design principles
elaborating on the functions, processes and interfaces of autonomic networks and systems.
The present document starts by analysising. The starting point of this work relies on the analysis of the existing research
initiatives, autonomic architectures and standards for telecommunications network operation and management. This
analysis allows to identify the gaps (or enablers to be introduced) between the current situation and the requirements
and expectations expressed by the actors of the telecommunications environment.
These requirements are detailed in the GS AFI 001 [i.51] Scenarios, Use Cases and Requirements for future self-
managing networks. In this way, the approach used to elaborate the present document is requirement-driven. In addition
to these requirements, the analysis allows to list the networking functions commonly needed or employed in the current
or emerging architectures, and identifying the missing functionalities.
The present document defines and describes what we call a Generic Autonomic Network Architecture (GANA)
Reference Model for Autonomic Network Engineering, Cognitive Networking and Self-Management. The present
document was produced by AFI WI#2 as shown on the figure below and subsequent releases of the Specification will
be produced and published as the Reference Model evolves. The Generic Autonomic Network Architecture (GANA)
Model is a Conceptual Architectural Reference Model for Autonomic Network Engineering, Cognition and Self-
Management. Its purpose is to serve as a "blueprint model" that prescribes the design and operational principles of
"autonomic decision-making manager components/elements" responsible for performing "autonomic" and "cognitive"
management and adaptive control of resources. It is not an implementation architecture per se. [An elaborate definition
is given in the next clauses]. Two aspects need to be distinguished as indicated on Figure 1:
1) a Generic Reference Model (specified in the present document); and
2) autonomicity-Enabled Reference Architectures that are the result of "instantiating" selected or all Functional
Blocks and Reference Points defined in the Reference Model onto a target implementation-oriented
standardized Reference Architecture. Clause 5 discusses the subject in detail.
In this first release of the GS AFI 002 (the present document), we aim to sketch a primary generic architecture model
definition, it is essential properties, its principles, its building blocks and their interactions. In this sense, this work may
be considered as a continuous process advocating for an incremental release of the AFI Generic Architecture reference
model intended to include enhancement, refinement and revision as the techniques mature, and contributions from the
community grow.
Figure 1 presents the current AFI Work Items and their relationships.
ETSI
9 ETSI GS AFI 002 V1.1.1 (2013-04)

Figure 1: Current AFI Work Items and their relationships
The present document provides the definition of an abstract reference representation (or architecture) for specifying
physical autonomic network elements and systems.
Individual Functional Blocks (and the associated principles) of said abstract representation could be seen as functional
elements, or architectural components, performing certain functions, and interfaces (through which they interact).
The Abstract Architectural Reference Model presented in the present document is described in a technology
independent way.
Characteristics of the Reference Model, including where reference points and governance issues are presented and
elaborated, what to standardize, etc, are all presented in the present document. The approach taken in the present
document is twofold:
1) Working out and Unifying viable Concepts and Design Principles for Autonomic Networking, Cognition and
Self-Management within a single Unified Framework (i.e. the Reference Model). In so doing, the present
document also includes the Techniques and Guidelines for addressing certain types of problems specific to the
process of designing, implementing, validating and deploying Autonomic Networks. Since the approach is "to
put ourselves in the shoes of the designer (primarily)", we seek to provide all necessary guidelines to the
industry w.r.t. to designing autonomic functions and incorporating them into the network and its elements.
2) Identifying the items that can be standardized as guided by the Reference Model for Autonomic & Cognitive
Networking and Self-Management, e.g. Functional Blocks (FBs) specific to realizing autonomicity, cognition
and self-management, Reference Points and Data Models describing the information/data communicated on
FBs' Interfaces, Autonomic Behaviours (i.e. control-loops) spanning multiple network elements, etc.
AFI is not only addressing the autonomic network management in the sense of automation of management processes as
well as introducing control-loops in the traditional Management Plane, but AFI takes a holistic approach to the problem
of "Self-Management and Adaptive Control" in the network elements and the network as a whole. This means there are
3 views to "management" that we consider and introduce Autonomics/Self-Management:
1) The traditional Vertical View that looks at the interface between a Network Element and the Management
System (EMS's/NMS's), as well as the whole Vertical Management Framework.
2) A Horizontal View of a "management-like" behaviour that may be called a "network-intrinsic" management or
"in-network management" that involves the collaboration of network elements along an E2E path (either hop-
by-hop or elements on some elements that are not necessarily on-link neighbours). It may simply be viewed as
an enhancement to the Control Plane with interacting distributed control-loops that enable network elements to
negotiate configurations and to adaptively control the behaviour or resources (protocols included) i.e. to
collaboratively self-optimize.
ETSI
10 ETSI GS AFI 002 V1.1.1 (2013-04)
3) The "device/element-intrinsic" self-management aspects i.e. autonomic functions introduced into the
architecture of a Network Element.
All the 3 views have some implications on Architectural Principles for both Evolved Networks and Future Network
Architectures that exhibit Self-Management Capabilities from the dawn of their design, and therefore the Concepts and
Principles defined by the Reference Model can also be applied to Future Network Architectures designed from scratch,
apart from being applicable to the evolution of existing network architectures towards "Autonomicity-Enabled
Architectures". The Reference Model presented in the present document covers all the 3 views listed above, in order to
capture a holistic picture on the evolution of management paradigms.
1.1 Reading Guide and Snapshots of the Reference Model
The following is a Reading Guide aimed at guiding readers (users) with different levels of knowledge and background
in Autonomics, Cognitive networking and Self-Management, and with diverse interests on various aspects of the
Reference Model and contents of the present document.
To be directed to specific chapters of interest for you the Reader (User) here is the guide:
1) The scope summarizes the scope and links to other documents or Work Items in AFI (Readers/Users: ALL are
recommended to read).
2) Readers/Users who are not familiar with the domain of Autonomic Networking, Cognition and
Self-Management and want to know about Characteristics and Properties of systems exhibiting
autonomic/self-management capabilities, can start their reading from clauses: 4. Definition of the AFI
Architectural Reference Model of a Generic Autonomic Network Architecture and its fundamental
requirements; 5. Properties of a Generic Autonomic Network Architecture Reference Model; 6. Enabling
concepts and mechanisms.
3) Readers/Users who are already familiar with Autonomic Networking, Cognition and Self-Management, and
simply want to focus on understanding the Reference Model can move on to clause 5 and then (just browse
through clause 8) or move straight to clauses 9 to 12, 13 and 14 that summarizes the various Reference Points
at the end. The snapshots of the Reference Model provided in this clause (see below) give a nutshell view that
should be enhanced by knowledge acquired in reading clauses 9 to 14 where Functional Blocks, Reference
Points and Principles are described in detail. Figures in the clauses 9 to 14 define the Reference Model. The
main figures are: Figure 2 to Figure 5; Figure 7 to Figure 11; Figure 14 to Figure 30; Figure 34, Figure 39,
Figure 49, Figure 63, Figure 64, Figure 67, Figure 68 and Figure 69.
4) Readers/Users who are already familiar with the domain of Autonomic Networking, Cognition and
Self-Management, and want to know how Concepts and Principles from various Models related to viable
approaches to the field, such as the IBM-MAPE model, FOCALE, 4D, CONMan, GENI, Knowledge Plane for
the Internet, etc., are all incorporated, harmonized and accommodated within GANA Reference Model as a
single Unifying Model, can move to clause 8 and also read the table in clause 10.12. The Table describes how
concepts from the different viable approaches to the domain have been selectively combined in a harmonized
way or accommodated within a single unifying model referred to as the GANA Reference Model in the
present document.
5) Readers/Users who want to know where "Place-holders for Controls-Loops" and Hierarchical and Horizontal
relationships/interfaces between them are defined, as well as where Cognition can be developed, can go
through the main figures of the Reference Model in clauses 9 to 11 (see also the snapshots in this clause as
well). Wherever there is a Decision-Element (DE), a Control-Loop for the DE can be designed. A "one-to-one"
mapping/assignment between specific DEs and the specific types of Managed Enties (MEs) and their
Configurable and Controllable Parameters over which a control-loop can be designed for a particular DE, is
provided in a table in clause 10. The notions of "fast control-loops" that can go into a Network Element (NE)
and "slower outer control-loops" are described in clause 9.
6) Readers/Users who want to know about Techniques and Guidelines for addressing certain types of problems
specific to designing, implementing, validating and deploying Autonomic Networks, such as how to address
Stability of control-loops, Trust and Confidence building by the operator can find details in clauses 10 and 11.
ETSI
11 ETSI GS AFI 002 V1.1.1 (2013-04)
7) Readers/Users who want to know about the Reference Model's mappings and implications to the TMN Logical
Layered Architecture (LLA), reflection of the FCAPS Framework within the Reference Model, and how the
Reference Model impacts or can be used to evolve/enhance existing EMSs/NMSs (OSS's) and other types of
implementation-oriented aspects can see details in clause 12. Topics such as Network Governance are also
covered in the corresponding clauses.
1.2 Snapshots of the Reference Model (a nutshell view)
The GANA (Generic Autonomic Network Architecture) Reference Model is a unified model for Autonomic
Networking, Cognition, and Self-Management. It defines generic Functional Blocks and associated Reference Points
and Characteristic Information that are specific to enabling autonomics, cognition, and self-management in a target
architecture. Therefore, it can be "instantiated" onto an implementation-oriented reference architecture such as the
3GPP architecture, BBF architecture, or ITU-T (NGN) architecture. The generic Functional Blocks and Reference
Points can also be applied in designing future network architectures that exhibit self-management capabilities from the
dawn (outset) of their design.
To what types of stakeholders is the Reference Model addressed to? Network architects, researchers, and
developers/implementers "refer" to the Reference Model when reasoning about or applying the concepts and principles
defining the domain of autonomic communication, autonomic networking, autonomic and cognitive management and
control-all as part of the "big-picture" of Self-Management.
Figures 2 and 3 provide just snapshots of the Reference Model for selected key aspects. Figure 2 provides an overview
of some of the key aspects of the GANA Reference Model. Figure 3 presents an instantiation of GANA Model within a
router, to illustrate instantiation (though not complete instantiation), with respect to autonomic routing as a working
example. Figures 2 and 3, are simplified versions of the more detailed Figure 34 in the core part of the present
document. In general, self-manageability in GANA is achieved through instrumenting the network elements (in Figure
3, the case of routers) with autonomic Decision-making-Elements (DEs) that collaboratively work together. DEs may
form "peers" along a path within the fundamental E2E transport architecture. The DE-2-DE peers need not necessarily
be hop-by-hop neighbours (i.e. being resident in on-link neighbouring nodes) but the peer relationships may relate to
e.g. border-relationships management in a heterogeneous network or may related to some DEs in certain network
elements along an E2E path. The Reference Model defines a hierarchy of DEs, i.e. four basic levels of self-
management: the protocol, function, node, and network levels. Each DE manages one or more lower-level DEs through
a control loop. These lower-level DEs are therefore considered Managed Entities (MEs). Over the control loop, the DE
sends commands, objectives, and policies to an ME and receives feedback in the form of monitoring information or
other type of knowledge. Each DE realizes some specific control-loop(s), and therefore, represents an "Autonomic
Activity" or "Autonomic Function" (AF). Examples of Autonomic Functions: Autonom
...