ETSI TR 103 747 V1.1.1 (2021-11)

Core Network and Interoperability Testing (INT/ WG AFI); Federated GANA Knowledge Planes (KPs) for Multi-Domain Autonomic Management & Control (AMC) of Slices in the NGMN(R) 5G End-to-End Architecture Framework

DTR/INT-00165

General Information

• Status: Not Published
• Technical Committee: INT AFI - Evolution of Management towards Autonomic Future Internet
• Current Stage: 12 - Completion
• Due Date: 15-Oct-2021
• Completion Date: 29-Nov-2021

Standards Content (Sample)

TECHNICAL REPORT
Core Network and Interoperability Testing (INT/WG AFI);
Federated GANA Knowledge Planes (KPs) for Multi-Domain
Autonomic Management & Control (AMC) of Slices in the ®
NGMN 5G End-to-End Architecture Framework

2 ETSI TR 103 747 V1.1.1 (2021-11)

Reference
DTR/INT-00165
Keywords
artificial intelligence, slicing

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

Tel.: [phone]  Fax: [phone]

Siret N° 348 623 562 00017 - APE 7112B
Association à but non lucratif enregistrée à la
Sous-Préfecture de Grasse (06) N° w061004871

Important notice
The present document can be downloaded from:
http://www.etsi.org/standards-search
The present document may be made available in electronic versions and/or in print. The content of any electronic and/or
print versions of the present document shall not be modified without the prior written authorization of ETSI. In case of any
existing or perceived difference in contents between such versions and/or in print, the prevailing version of an ETSI
deliverable is the one made publicly available in PDF format at www.etsi.org/deliver.
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
https://portal.etsi.org/TB/ETSIDeliverableStatus.aspx
If you find errors in the present document, please send your comment to one of the following services:
https://portal.etsi.org/People/CommiteeSupportStaff.aspx
Notice of disclaimer & limitation of liability
The information provided in the present deliverable is directed solely to professionals who have the appropriate degree of
experience to understand and interpret its content in accordance with generally accepted engineering or
other professional standard and applicable regulations.
No recommendation as to products and services or vendors is made or should be implied.
No representation or warranty is made that this deliverable is technically accurate or sufficient or conforms to any law
and/or governmental rule and/or regulation and further, no representation or warranty is made of merchantability or fitness
for any particular purpose or against infringement of intellectual property rights.
In no event shall ETSI be held liable for loss of profits or any other incidental or consequential damages.

Any software contained in this deliverable is provided "AS IS" with no warranties, express or implied, including but not
limited to, the warranties of merchantability, fitness for a particular purpose and non-infringement of intellectual property
rights and ETSI shall not be held liable in any event for any damages whatsoever (including, without limitation, damages
for loss of profits, business interruption, loss of information, or any other pecuniary loss) arising out of or related to the use
of or inability to use the software.
Copyright Notification
No part may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and
microfilm except as authorized by written permission of ETSI.
The content of the PDF version shall not be modified without the written authorization of ETSI.
The copyright and the foregoing restriction extend to reproduction in all media.

© ETSI 2021.
All rights reserved.
ETSI
3 ETSI TR 103 747 V1.1.1 (2021-11)
Contents
Intellectual Property Rights . 5
Foreword . 5
Modal verbs terminology . 5
Introduction . 5
1 Scope . 6
2 References . 6
2.1 Normative references . 6
2.2 Informative references . 6
3 Definition of terms, symbols and abbreviations . 9
3.1 Terms . 9
3.2 Symbols . 9
3.3 Abbreviations . 9
4 Principles for Autonomic Networking and Autonomic Management & Control (AMC), and
Enablers . 11
4.1 Overview on Autonomics Principles and Enablers for Autonomic Networking, Autonomic Management
& Control (AMC), and Autonomous Networks . 11
4.2 Closed Control-Loop(s) . 13
4.3 Introduction to the ETSI GANA Reference Model for Autonomic Networking, Cognitive Networking
and Self-Management . 14
4.4 Federation of GANA Knowledge Planes Framework . 22 ®
4.5 AMC Requirements in the NGMN 5G E2E Architecture, and need for Knowledge Plane Federations ®
for E2E AMC in NGMN E2E 5G Architecture . 23
4.6 The Value of Autonomics in Network Slicing . 24
5 SliceNet Architecture . 25
5.1 Overview . 25
5.2 Control Framework . 26
5.3 Cognitive management . 28
5.3.0 Introduction. 28
5.3.1 Cognitive Control Loop . 29
5.3.2 Knowledge & Monitoring . 30
5.3.3 Analysis . 30
5.3.4 Planning & Execution . 33
5.4 Slice management. 35
5.5 Orchestration . 36
6 Impact of MEC, Network Slicing and Hardware Acceleration to the SliceNet Concepts and
Principles . 37
6.1 Impact of Virtualization . 37
6.2 Impact of MEC . 38
6.3 Impact of Network Slicing . 39
6.4 Impact of Hardware Acceleration. 40
7 GANA in ETSI 5G PoC Implementations by the Industry . 41
8 Mapping of SliceNet architecture components to GANA Concepts and Architectural Principles,
How to use the SliceNet components to implement GANA Components . 52
8.1 General Mapping of SliceNet Architectural Concepts and Principles to GANA Concepts and Principles . 52
8.2 Autonomic networks and General GANA integration with SDN, NFV, Data Analytics Applications,
Orchestrators, and Other Management and Control Systems . 53
8.3 SliceNet mapping to GANA Network Level (Knowledge Plane (KP) Level) Autonomics . 55
8.4 How to implement a GANA Knowledge Plane (KP) for a specific network segment using the SliceNet
Intelligence Framework . 56 ®
9 Addressing the AMC Requirements in the NGMN 5G E2E Architecture . 57
ETSI
4 ETSI TR 103 747 V1.1.1 (2021-11)
10 Conclusion . 58
Annex A: Bibliography . 60
History . 61

ETSI
5 ETSI TR 103 747 V1.1.1 (2021-11)
Intellectual Property Rights
Essential patents
IPRs essential or potentially essential to normative deliverables may have been declared to ETSI. The declarations
pertaining to these essential IPRs, if any, are 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 (https://ipr.etsi.org/).
Pursuant to the ETSI Directives including the ETSI IPR Policy, no investigation regarding the essentiality of IPRs,
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.
Trademarks
The present document may include trademarks and/or tradenames which are asserted and/or registered by their owners.
ETSI claims no ownership of these except for any which are indicated as being the property of ETSI, and conveys no
right to use or reproduce any trademark and/or tradename. Mention of those trademarks in the present document does
not constitute an endorsement by ETSI of products, services or organizations associated with those trademarks.
DECT™, PLUGTESTS™, UMTS™ and the ETSI logo are trademarks of ETSI registered for the benefit of its

Members. 3GPP™ and LTE™ are trademarks of ETSI registered for the benefit of its Members and of the 3GPP
Organizational Partners. oneM2M™ logo is a trademark of ETSI registered for the benefit of its Members and of the ®
oneM2M Partners. GSM and the GSM logo are trademarks registered and owned by the GSM Association.
Foreword
This Technical Report (TR) has been produced by ETSI Technical Committee Core Network and Interoperability
Testing (INT).
Modal verbs terminology
In the present document "should", "should not", "may", "need not", "will", "will not", "can" and "cannot" are to be
interpreted as described in clause 3.2 of the ETSI Drafting Rules (Verbal forms for the expression of provisions).
"must" and "must not" are NOT allowed in ETSI deliverables except when used in direct citation.
Introduction
The present document provides a mapping and evaluation of architectural components for autonomic network
management & control developed/implemented in the EU-funded SliceNet Project to the ETSI AFI Generic Autonomic
Networking Architecture (GANA) model - an architectural reference model for autonomic networking, cognitive
networking and self-management. It serves to provide useful insights to implementers of ETSI GANA Knowledge
Plane (KP) Platforms on approaches that can be taken in implementing ETSI GANA Knowledge Plane Platforms and
how to federate them for E2E (Cross-Domain) Autonomic Management and Control (AMC) operations across network
segments such as Radio Access Networks (RANs), Transport Networks and Core Networks. It goes further to discuss ®
how to address the AMC Requirements specified in NGMN E2E 5G Architecture by a way of providing insights on
how ETSI GANA KP platforms in an E2E 5G Architecture can address the AMC requirements, while leveraging
experiences gained in SliceNet Project in implementing GANA autonomic/cognitive management and control software
components for 5G network slices.

ETSI
6 ETSI TR 103 747 V1.1.1 (2021-11)
1 Scope
The present document presents a plausible approach to implementing Federated GANA Knowledge Planes (KPs)
Platforms for E2E Multi-Domain Federated Autonomic Management and Control (AMC) of 5G Network Slices in ®
NGMN E2E 5G Architecture, using components prototyped and implemented in the European Union (EU) funded
SliceNet Project (Grant Agreement N° 761913). The present document produces and leverages a mapping of
architectural components for autonomic network management & control developed/implemented SliceNet Project to the
ETSI TC INT AFI Generic Autonomic Networking Architecture (GANA) model - an architectural reference model for
autonomic networking, cognitive networking and self-management. The mapping identifies the components that were
prototyped in Slicenet Project that can be used to implement specific GANA Functional Blocks (FBs) for Autonomics
and their associated Reference Points (Rfps), while providing the illustrations that help implementers of GANA
autonomics in 5G networks. Other aspects covered in the present document are: ®
• A Study of GANA aligned AMC Requirements in the NGMN 5G E2E Architecture in order to provide
answers on how the approach presented in the present document can help implementers of GANA AMC
solutions for 5G.
• Providing useful insights to implementers of ETSI GANA Knowledge Plane (KP) Platforms on approaches
that can be taken in implementing ETSI GANA Knowledge Plane Platforms and how to federate them for E2E
(Cross-Domain) Autonomic Management and Control (AMC) operations across network segments such as
Radio Access Networks (RANs), Transport Networks and Core Networks.
• Providing insights on leveraging experiences gained in SliceNet Project in implementing GANA
autonomic/cognitive management and control software components for 5G network slices.
The mapping of the components to the GANA model concepts serves to illustrate how to implement the key abstraction
levels for autonomics (self-management functionality) in the GANA model for the targeted wireless networks context,
taking into consideration the work done in ETSI TR 103 495 [i.26].
It also shows how GANA can be implemented using the components developed in SliceNet project as an example.
2 References
2.1 Normative references
Normative references are not applicable in the present document.
2.2 Informative references
References are either specific (identified by date of publication and/or edition number or version number) or
non-specific. For specific references, only the cited version applies. For non-specific references, the latest version of the
referenced document (including any amendments) applies.
NOTE: While any hyperlinks included in this clause were valid at the time of publication, ETSI cannot guarantee
their long term validity.
The following referenced documents are not necessary for the application of the present document but they assist the
user with regard to a particular subject area.
[i.1] ETSI White Paper No.16 GANA: "Generic Autonomic Networking Architecture Reference Model
for Autonomic Networking, Cognitive Networking and Self-Management of Networks and
Services".
NOTE: Available at http://www.etsi.org/images/files/ETSIWhitePapers/etsi_wp16_gana_Ed1_20161011.pdf.
ETSI
7 ETSI TR 103 747 V1.1.1 (2021-11)
[i.2] ETSI TS 103 195-2 (V1.1.1): "Autonomic network engineering for the self-managing Future
Internet (AFI); Generic Autonomic Network Architecture; Part 2: An Architectural Reference
Model for Autonomic Networking, Cognitive Networking and Self-Management".
NOTE: Available at https://portal.etsi.org/webapp/WorkProgram/Report_WorkItem.asp?WKI_ID=50970.
[i.3] ETSI 5G PoC on 5G Network Slices Creation, Autonomic & Cognitive Management & E2E
Orchestration-with Closed-Loop (Autonomic) Service Assurance for the IoT (Smart Insurance)
Use Case.
NOTE: More information at https://intwiki.etsi.org/index.php?title=Accepted_PoC_proposals.
[i.4] White Paper No.1 of the ETSI 5G PoC: "C-SON Evolution for 5G, Hybrid SON Mappings to the
ETSI GANA Model, and achieving E2E Autonomic (Closed-Loop) Service Assurance for 5G
Network Slices by Cross-Domain Federated GANA Knowledge Planes".
NOTE: More information at
https://intwiki.etsi.org/images/ETSI_GANA_in_5G_PoC_White_Paper_No_1_v1.28.pdf.
[i.5] ETSI TR 103 473 (V1.1.2): "Evolution of management towards Autonomic Future Internet (AFI);
Autonomicity and Self-Management in the Broadband Forum (BBF) Architectures".
NOTE: Available at
https://www.etsi.org/deliver/etsi_tr/103400_103499/103473/01.01.02_60/tr_103473v010102p.pdf.
[i.6] ETSI TR 103 404: "Network Technologies (NTECH); Autonomic network engineering for the
self-managing Future Internet (AFI); Autonomicity and Self-Management in the Backhaul and
Core network parts of the 3GPP Architecture". ®
Open Source Project: "ONAP Architecture Overview".
[i.7] ONAP
NOTE 1: Available at https://www.onap.org/. ®
NOTE 2: Linux is the registered trademark of Linus Torvalds in the U.S. and other countries.
[i.8] BBF CloudCO Open Source Project.
NOTE: Available at https://www.broadband-forum.org/cloudco. ®
[i.9] OPNFV Open Source Project.
NOTE 1: Available at https://www.opnfv.org/. ®
NOTE 2: Linux is the registered trademark of Linus Torvalds in the U.S. and other countries.
[i.10] ONOS Open Source Project.
NOTE: Available at https://onosproject.org/. ®
[i.11] OpenDayLight Open Source Project.
NOTE 1: Available at https://www.opendaylight.org/. ®
NOTE 2: Linux is the registered trademark of Linus Torvalds in the U.S. and other countries.
[i.12] ETSI OSM (Open Source MANO).
NOTE: Available at https://osm.etsi.org/. ®
[i.13] ACUMOS : "An Open Source AI Machine Learning Platform".
NOTE 1: Available at https://www.acumos.org/. ®
NOTE 2: ACUMOS is a registered trademark of LF Projects, LLC.
ETSI
8 ETSI TR 103 747 V1.1.1 (2021-11)
[i.14] White Paper No.3 of the ETSI 5G PoC: "Programmable Traffic Monitoring Fabrics that enable
On-Demand Monitoring and Feeding of Knowledge into the ETSI GANA Knowledge Plane for
Autonomic Service Assurance of 5G Network Slices; and Orchestrated Service Monitoring in
NFV/Clouds".
NOTE: Available at ETSI_5G_PoC_White_Paper_No_3_2019_v1.19.pdf.
[i.15] White Paper No.2 of the ETSI 5G PoC: "ONAP Mappings to the ETSI GANA Model; Using
ONAP Components to Implement GANA Knowledge Planes and Advancing ONAP for
Implementing ETSI GANA Standard's Requirements; and C-SON - ONAP Architecture".
NOTE: Available at ETSI_5G_PoC_White_Paper_No_2_Final_v7.3.pdf.
[i.16] ETSI 5G PoC Report on Specifications of Integration APIs for the ETSI GANA Knowledge Plane
Platform with Other Types of Management & Control Systems, and with Info/Data/Event Sources
in general.
NOTE: Available at https://intwiki.etsi.org/index.php?title=Accepted_PoC_proposals.
[i.17] ETSI TS 129 520 (V16.6.0): "5G; 5G System; Network Data Analytics Services; Stage 3 (3GPP
TS 29.520 version 16.6.0 Release 16)".
[i.18] ETSI TS 128 533 (V15.0.0): "5G; Management and orchestration; Architecture framework (3GPP
TS 28.533 version 15.0.0 Release 15)". ®
[i.19] 5G End-to-End Architecture Framework by NGMN Alliance: "P1-Requirements and ®
Architecture: NGMN 5G E2E Architecture Framework v3.0.8".
NOTE: Available at https://www.ngmn.org/publications/5g-end-to-end-architecture-framework-v3-0-8.html.
[i.20] White Paper No.4 of the ETSI 5G PoC: "ETSI GANA as Multi-Layer Artificial Intelligence (AI)
Framework for Implementing AI Models for Autonomic Management & Control (AMC) of
Networks and Services; and Intent-Based Networking (IBN) via GANA Knowledge Planes
(KPs)".
NOTE: Available at ETSI_5G_PoC_White_Paper_No_4_v3.1.pdf.
[i.21] White Paper No.6 of the ETSI 5G PoC: "Generic Framework for Multi-Domain Federated ETSI
GANA Knowledge Planes (KPs) for End-to-End Autonomic (Closed-Loop) Security Management
& Control for 5G Slices, Networks/Services".
NOTE: Available at ETSI_5G_PoC_White_Paper_No_6.pdf.
[i.22] ETSI GS AFI 002 (V1.1.1): "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)".
[i.23] SliceNet Project Deliverable D5.7: "Framework for Cognitive SLA and QoE Slice Management",
December 2019.
NOTE: Available at https://doi.org/10.18153/SLIC-761913-D5_7.
[i.24] SliceNet Project Deliverable D5.5: "Modelling, Design and Implementation of QoE Monitoring,
Analytics and Vertical-Informed QoE Actuators, Iteration I.
NOTE: Available at https://doi.org/10.18153/SLIC-761913-D5_5.
[i.25] SliceNet Project Deliverable D5.6: "Modelling, Design and Implementation of QoE Monitoring,
Analytics and Vertical-Informed QoE Actuators, Iteration II".
NOTE: Available at https://doi.org/10.18153/SLIC-761913-D5_6.
[i.26] ETSI TR 103 495: "Network Technologies (NTECH); Autonomic network engineering for the
self-managing Future Internet (AFI); Autonomicity and Self-Management in Wireless Ad-
hoc/Mesh Networks: Autonomicity-enabled Ad-hoc and Mesh Network Architectures".
ETSI
9 ETSI TR 103 747 V1.1.1 (2021-11)
[i.27] White Paper No.5: "Artificial Intelligence (AI) in Test Systems, Testing AI Models and ETSI
GANA Model's Cognitive Decision Elements (DEs) via a Generic Test Framework for Testing
GANA Multi-Layer Autonomics & their AI Algorithms for Closed-Loop Network Automation".
NOTE: Available at ETSI_5G_PoC_White_Paper_No_5.pdf.
3 Definition of terms, symbols and abbreviations
3.1 Terms
Void.
3.2 Symbols
Void.
3.3 Abbreviations
For the purposes of the present document, the following abbreviations apply:
rd
3GPP 3 Generation Partnership Project
AFI Autonomic Future Internet
AIOPS Artificial Intelligence for IT Operations
AMC Autonomic Management & Control
AMF Access and Mobility Management Function
AN Autonomous Network
API Application Programming Interface
AUSF AUthentication Server Function
BBF BroadBand Forum
CN Core Network
CP Control Plane
CPS Control Plane Services
CPSR Control Plane Service Register
CPU Central Processing Unit
CRUD Creation/Configuration, Read, Update and Delete
C-SON Centralized Self Organizing Network
CSP Communications Service Provider
DB DataBase
DDCM Data-Driven Control and Management
DE Decision making Element
DP Data Plane
DSON Distributed SON
DSP Digital Service Provider
E2E End-to-End
EMS Element Management System
EPC Evolved Packet Core
EPS Edge Packet Service
FCAPS Fault, Configuration, Accounting, Performance, Security
FGE Forward Graph Enabler
GANA Generic Autonomic Network Architecture
GW-C GateWay-Control plane
GW-U GateWay-User plane
IBN Intent-Based Networking
ICT Information and Communications Technology
IMSI International Mobile Subscriber Identity
IP Internet Protocol
IPC Inter PoP Connection
ETSI
10 ETSI TR 103 747 V1.1.1 (2021-11)
ISG Industry Specification Group
IT Information Technology
JSON Java Script Notation Object
KB Knowledge Base
KP DE Knowledge Plane Decision-making Element
KP Knowledge Plane
KPI Key Performance Indicator
LL-MEC Low-Latency MEC
MANO MANagement and Orchestration
MAPE-K Monitor-Analyze-Plan-Execute over a shared Knowledge
MB MegaByte
MBTS Model-Based Translation Service
MDAS Management Data Analytics Service
MEC Mobile Edge Computing
ML Machine Learing
NBI Northbound interface
NE Network Element
NF Network Function
NFV Network Function Virtualisation
NFVO NFV Orchestrator ®
Next Generation Mobile Networks
NGMN
NMR-O NFV MEC RAN Orchestrator
NRF Network Repository Function
NS Network Slice
NSO Network Service Orchestrator
NSP Network Service Provider
NSS Network Sub-Slice
NSSF Network Slice Selection Function
NSST Network Slice Subnet Templates
NST etwork Slice Template
NWDAF Network Data Analytics Function
OAI OpenAirInterface
ONAP Open Network Automation Platform
ONIX Overlay Network for Information eXchange
OOB Out-Of-Band
OODA Observe-Orient-Decide-ACT
OSA One Stop API
OSM Open Source MANO
OSS Operations Support Systems
OVS Open Virtual Switch
P&P Plug & Play
PAP Policy Administration Point
PCF Policy Control Function
PCI Policy Catalogue & Inventory
PCM Policy Control Manager
PCS Proactive Control Scheme
PDP Policy Decision Point
PF Policy Framework
PGW Packet Gateway
PNF Physical Network Function
PR Policy Recommender
QoE Quality of Experience
QoS Quality of Service
RAN Radio Access Network
REST Representational State Transfer
RNIS Radio Network Information Service
SBA Service Based Architecture
SBI Service Based Interface
SDK Software Development Kit
SDN Software Defined Networks
SDO Standards Development Organizations
SGW Serving GateWay
ETSI
11 ETSI TR 103 747 V1.1.1 (2021-11)
SLA Service Level Agreement
SMF Session Management Function
SNMP Simple Network Management Protocol
SON Self Organizing Networks
SS-O Service and Slice Orchestrator
TAL Tactical Autonomic Language
TCAM Ternary Content Addressable Memories
UDM Unified Data Management
UE User Equipment
UI User Interface
UP User Plane
UPF User Plane Function
VIM Virtual Infrastructure Management
VLAN Virtual LAN
VNF Virtual Network Function
WAN Wide Area Network
WG Working Group
WIM WAN Infrastructure Manager
4 Principles for Autonomic Networking and Autonomic
Management & Control (AMC), and Enablers
4.1 Overview on Autonomics Principles and Enablers for
Autonomic Networking, Autonomic Management & Control
(AMC), and Autonomous Networks
Autonomic systems rely on an autonomic elements which regularly sense the possible sources of change through
sensors, reason about the current situation and arrange adaptations through actuators. Autonomous networks focuses on
the definition of closed loops (MAPE-K, OODA, cognitive loop, etc.) or controllers as enablers for autonomy in future
networks. The Autonomous Networks aims to define fully automated innovative network and services for vertical
industries' users and consumers, supporting self-configuration, self-healing, self-optimizing and self-evolving network
infrastructures. Autonomous Networks incorporate a simplified network architecture, autonomous domains and
automated intelligent business/network operations for the closed control loop, offering the best-possible user
experience, full lifecycle operations automation/autonomy and maximum resource utilization. One of the Key
inseparable features of an autonomic system is a need of continuous monitoring.
Self-manageability in GANA is achieved through instrumenting the network with autonomic Decision-making-
Elements (DEs), which automate network operations by implementing control loops. Such control loops operate using
the knowledge regarding events and the state of network resources.
The GANA model defines a generic Autonomic Management and Control (AMC) framework and structure within
which to specify and design autonomics-enabling functional blocks for any network architecture and its management
architecture. Autonomic Management & Control is about Decision making elements (Des) as autonomic functions with
cognition introduced in control and management plane. Cognition is seen as learning and reasoning used to effect
advanced adaptation in Decision making Elements. Control is about control-logic as the core of DE that realizes a
control-loop in order to adjust network resources/parameters or services. From an architecture perspective, AMC
Framework and 3GPP Hybrid-SON (Self Organizing Network) model are compatible with each other. Both shares
common design principles on enabling implementers of autonomics algorithms to combine centralized and distributed
control of network resources, parameters and services. Federation of AMC allows knowledge exchanging, sharing,
interaction and collaboration among KPs with each KP platform governing and controlling the behaviour of an
autonomic system.
ETSI
12 ETSI TR 103 747 V1.1.1 (2021-11)
The following concepts and paradigms help implementers of ETSI GANA autonomics in understanding important
taxonomy of terms and concepts that related in this area of the drive to smart and self-driving networks:
• Autonomic Network (AcN)--concept: An Autonomic Network either refers to the Network Infrastructure
made up of Network Elements/Functions (NEs/NFs) that exhibit autonomics (control-loops) in their behaviors,
or is the inclusion together of Network Infrastructure with some level of autonomics (control-loops) in
NEs/NFs and its associated Management and Control architecture that also exhibits autonomics (Control-
Loops) at that higher level. A Multi-Layer AcN's property of being "autonomic" (also called "autonomicity"
(a paradigm) in ETSI Standards such as [i.2]), relates to Hierarchical (nested) and Interworking Control-
Loops introduced at various Abstraction Levels (from NE/NF level) up into the Management & Control
Systems Level of the Network.
NOTE: On the journey to implementing an Autonomous Network (AN) with certain degree of autonomy, the
starting point is designing the network as an Autonomic Network (AcN) as the foundation. The science of
Control-Loops (called "Autonomics") is key enabler for achieving the property of being autonomous, and
that an AcN is expected to evolutionarily "maximize" the property of being "autonomous" in as far as the
"Degree and Measure of Operations Tasks that can be performed by the autonomic network (AN)
without direct human involvement in the decision and actions".
• An Autonomous Network (AN)--concept: An AN is a network that exhibits a property called Degree of
Autonomy that pertains to the Degree and Measure of Operations Tasks that can be performed by an
Autonomic Network (AcN) without direct human involvement in the decision and actions, thanks to
Autonomics (the science of Control-Loops) in-built in the AcN as the enablers for the autonomous behavior
(an operational property of the autonomic network). The Degree of Autonomy is associated with Maturity
Levels for Autonomy that are increasingly attained by the Autonomic Network (AcN) over time, thanks to the
Evolution of the Autonomics (Control-Loops) by enrichment of automation in network and services
management and control intelligence in the Control-Loops to maximize the autonomic network's property of
being Autonomous. In ICT networks, AN is to be governable by Human Operator through inputs such as
business goals and policies and other governance inputs such an Intents, Service Level Agreements (SLAs) for
services to be delivered by the AN.
• Automated Management (a paradigm): It is about workflow reduction and automation i.e. automation of the
processes involved in the creation of network and/or service configuration inputs using specialized task automation
tools (e.g. scripts, automated workflow management tools, network planning tools, policy generators for
conflict-free policies, intents, goals, Service Level Agreement (SLAs), etc.) such that the Inputs can be provided
by the Human Operator to the Autonomic Network (AcN) or Autonomous Network (AN) to govern its operation.
The Inputs (high-level inputs), also called Governance Objectives/Goals for the AcN/AN are further operated upon
by the AcN/AN to further derive detailed low-level configuration data and actions that are then applied by the
AcN/AN on its own to self-configure and execute other self-* operations such as self-optimization, self-diagnosis,
self-repairing/healing, self-protection during its operation. The AcN/AN exposes an interface called the
Governance Interface through which the Automated Management realm provides the Governance Inputs
(Objectives/Goals) to the AcN/AN. The same Governance Interface of the AcN/AN is used by the AcN/AN to
provide feedback to the Human (through Tools of the Automated Management realm) in form of Reports that
provide insights such as how the AcN/AN is fairing in fulfilling the Objectives, or feedback in form escalations that
the AcN/AN would like the Human Operator to get involved in the decisions on how to handle certain situations the
AcN/AN has encountered.
ETSI
13 ETSI TR 103 747 V1.1.1 (2021-11)
• Autonomic Management and Control (AMC) -a paradigm: It emphasizes learning, reasoning, and
adaptation using control-loops that also take into consideration the feedback knowledge obtained from
network and services monitoring. Automated Management provides Input (Governance Inputs) to the
Autonomic Management & Control (AMC) of Networks and Services Domain ("area/space"). [i.1] and [i.2]
define the AMC paradigm as the interworking of nested and hierarchical control-loops and associated logics
introduced in the Management Plane, in the Control Plane, and also in the converged (non-disaggregated)
Management Planes and Control Planes (as there are such cases). Indeed, Autonomic Management & Control
should exhibit a network Governance Interface through which the input that governs the configuration of an
Autonomic Network (AcN) should be provided by the human operator. Thanks to automation tools and
mechanisms (Automated Management), by using a high-level language, the operator can define the features of
the network services that should be provided by the underlying network infrastructure. Such a business
language that can help the operator express high level business goals required of the network may be modelled
by the use of an ontology to add semantics and enable machine reasoning on the goals. The human operator
defined features relate to business goals, technical goals and some input configuration data that an autonomic
network is supposed to use as operational targets (which may flexibly be changed or modified by the operator
at any time) for network resources and parameters configurations.
• In the relationship of the two properties of an AcN/AN (of being "autonomic" and being "autonomous"), one
can realize that the science of Control-Loops (called "Autonomics") is key enabler for achieving the property
of being autonomous, and that an AcN is expected to evolutionarily "maximize" the property of being
"autonomous" in as far as the "Degree and Measure of Operations Tasks that can be performed by the
Autonomic Network (AcN) without direct human involvement in the decision and actions". That means
that evolving the Autonomics (Control-Loops) of the AcN by enrichment of automation in management,
control and associated intelligence in the Control-Loops, Maximizes the AcN's property of being Autonomous
(but while still remaining governable and controllable by humans-particularly for telecommunication and IT
networks meant to serve business customers of network providers). Two Dimensions for Autonomics
Evolution of the AcN are:
1) Market Place Driven Evolution of Decision-making-Elements (DEs) for Autonomics enables onboarding
better DEs with better Algorithms;
2) Algorithmic Evolution of DEs' AI Algorithms for Autonomics.
The "evolution" takes different paces in deployed AcNs of various organizations. Therefore, Evolutionary
Autonomic Networks are characterized by the ability to evolve in maximization of the property of being
"Autonomous" (degree of Autonomy in operations tasks) by enriched automation using Control-Loops, to
reach an extent by which the human operators see themselves focused mainly on providing Governance Inputs
(such as Business Objectives, SLAs, Goals, Policies and Intents) to the Network while experiencing a drastic
reduction of involvement in any burdens involving tasks such as security management/control,
fault-management and Network Optimization processes for the network (thanks to autonomics features such as
self-repair, self-healing, self-protection and self-optimization, self-awareness, by the network on its own).
For more details on taxonomy harmonization in this space and enablers for the paradigms, the following sources
provide useful insights [i.20], [i.1] and [i.2].
4.2 Closed Control-Loop(s)
Traditional networks based on human labour-intensive, time and resource-consuming management tasks, carried out by
operations teams, should be conducted autonomously by following a much more efficient, intelligent and automated
machine-oriented approach. Network transformation and the evolution from the traditional communications industry to
the digital services industry demands a paradigm shift in the way networks are planned, deployed and operated.
The evolution to the digital services industry imposes additional challenges on the agility, flexibility and robustness
needed to manage the lifecycle of services and underlying resources. In the digital services industry, service innovation
cycles are becoming much shorter, service activation is expected to be instantaneous, and services are supposed to be
always available and highly responsive. Due to these requirements, the digital services industry is incompatible with
human dependencies on service activation, optimization and recovery situations.
ETSI
14 ETSI TR 103 747 V1.1.1 (2021-11)
The ongoing network transformation towards Network Functions Virtualisation (NFV) and Software-Defined Networks
(SDNs) is becoming a reality and will result in a significantly improve in agility, flexibility and cost efficiency to
manage network functions, which are the foundations to trigger a paradigm shift in the way network operations are
planned, deployed and managed, called cognitive network management. This approach consists of implementing
machine-based intelligence to support the creation of autonomous processes to manage complex networking scenarios.
One of the main impacts of this paradigm is the significant reduction of operational costs. Proactive and reactive actions
are automated to resolve or mitigate networking problems, thereby minimizing the human effort in maintenance and
troubleshooting tasks, and leading to significant Operational Expenditure (OPEX) reduction.
Three key capabilities is to be provided to create a closed control loop solution able to support the implementation of
autonomous and intelligent cognitive processes:
• Automated network monitoring: the key challenge is to build transversal and automated monitoring
capabilities crossing all network domains. These can be achieved through the automated deployment of
virtualized probes in the network infrastructure to facilitate system-wide distributed monitoring. Collected
information has to feed data analysis algorithms, like data analytics, data mining or Machine Learning (ML),
in order to create key indicators that may translate to:
1) service affecting conditions (e.g. network failures, performance bottlenecks, security breaches,
intrusions);
2) conditions that may evolve to service affecting issues in the future;
3) non-optimal service delivery to specific users, i.e., detection of cases where the service topology being
used to deliver a service to end users can be optimized in order to minimize allocated resources or the
service QoS.
• Cognitive framework: the ability to define high level tactical corrective and preventive measures to respond
to the diagnosed conditions. Tactical measures may correspond to reactive actions to fix or mitigate existing
network issues or may correspond to proactive actions to prevent the evolution of the diagnosed condition to
an effective service-affecting anomaly. These actions typically correspond to services and/or network
functions lifecycle management requests (e.g. automated instantiation, configuration, scalability,
reconfiguration of connectivity logical topology, etc.).
• Automated and dynamic service provisioning: automated and intelligent processes to manage the lifecycle
of services and network functions. These comprise the dynamic selection of the best locations and resources
for services deployment (or migration) considering the requirements associated with the specific service
instance being provisioned (for instance the contracted QoS). This process also includes the provisioning of
the key performance indicators to be produced for the service instance.
4.3 Introduction to the ETSI GANA Reference Model for
Autonomic Networking, Cognitive Networking and
Self-Management
This clause introduces the ETSI GANA Reference Model to provide a basis for the objectives described in the scope of
the present document.
The ETSI Generic Autonomic Networking Architecture (GANA) Reference Model is an Architectural Reference Model
for Autonomic Networking, Cognitive Networking and Self-Management of Networks and Services standardized by
ETSI [i.1], [i.2], [i.22], which defines a generic Functional Blocks (FBs), their associated reference points, and
messages passed through those reference points. Figure 1 presents the snapshot of the ETSI GANA Reference Model
and the aspect of Multi-Layer Autonomics' Cognitive Algorithms for Artificial Intelligence (AI) and the levels of
abstractions of self-management functionality. Self-management functionality is a logic (component) that implements a
control-loop as the core driver of the self-management behavior in terms of orchestration and/or (re)-configuration of
entities that need to be orchestrated, managed and
...