Contents

1 Overview of Network Management 5

1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.1.1 Common Network Problems . . . . . . . . . . . . . . . . . 7
1.1.2 Challenges of IT Managers . . . . . . . . . . . . . . . . . 8
1.2 Network Management: Goals, Organization and Functions . . . . 9
1.2.1 Goal of Network Management . . . . . . . . . . . . . . . . 9
1.2.2 Various Functions of Network Management . . . . . . . . 11
1.3 Network Management Architecture . . . . . . . . . . . . . . . . . 14

2 OSI Network Management 19

2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.2 Network Management Standards . . . . . . . . . . . . . . . . . . 20
2.3 Network Management Models . . . . . . . . . . . . . . . . . . . . 24
2.4 Organization Model . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.5 Information Model . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.5.1 Management Information Tree and Base . . . . . . . . . . 29
2.6 Communication Model . . . . . . . . . . . . . . . . . . . . . . . . 30
2.7 Functional Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
2.8 Abstract Syntax Notation One (ASN.1) . . . . . . . . . . . . . . 34
2.8.1 Module . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

3 Internet Management (SNMP) 37

3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
3.2 System Overview : The SNMP Model . . . . . . . . . . . . . . . 39
3.3 Organization Model of SNMP . . . . . . . . . . . . . . . . . . . . 43
3.4 Communication Model . . . . . . . . . . . . . . . . . . . . . . . . 46
3.4.1 SNMP Architecture . . . . . . . . . . . . . . . . . . . . . 46
3.4.2 Administration Model of SNMP . . . . . . . . . . . . . . 48
3.4.3 SNMP Protocol Specifications . . . . . . . . . . . . . . . . 50
3.4.4 Flow of SNMP PDU . . . . . . . . . . . . . . . . . . . . 54
3.5 SNMP Proxy Server . . . . . . . . . . . . . . . . . . . . . . . . . 55
3.6 Remote Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . 56
3.6.1 RMON Advantages . . . . . . . . . . . . . . . . . . . . . . 57
3.6.2 RMON Structure of MIB . . . . . . . . . . . . . . . . . . 58

1
CONTENTS 2

3.7 RMON Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

3.8 SNMP Management Components . . . . . . . . . . . . . . . . . . 61
3.8.1 Functions of SNMP Components . . . . . . . . . . . . . . 61
3.9 TLV Encoding Method . . . . . . . . . . . . . . . . . . . . . . . . 63
3.9.1 Encoding Integer Data Type . . . . . . . . . . . . . . . . 63
3.9.2 Encoding Characters (OCTET STRING) . . . . . . . . . 64
3.9.3 Encoding IP Address . . . . . . . . . . . . . . . . . . . . . 64
3.10 Comparison Between Different SNMP Versions . . . . . . . . . . 65
3.11 SNMP Version 2: SNMPv2 . . . . . . . . . . . . . . . . . . . . . 65
3.11.1 SNMPv2: Improvements Over SNMPv1 . . . . . . . . . . 66
3.11.2 SNMPv2 System Architecture . . . . . . . . . . . . . . . . 67
3.11.3 Compatibility with SNMPv1: Bilingual Manger . . . . . . 69
3.12 SNMP Version 3: SNMPv3 . . . . . . . . . . . . . . . . . . . . . 70
3.12.1 SNMPv3 Architecture . . . . . . . . . . . . . . . . . . . . 70

4 Broadband Network Management 75

4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
4.2 Asynchronuous Transfer Mode (ATM) . . . . . . . . . . . . . . . 76
4.2.1 ATM Terminologies . . . . . . . . . . . . . . . . . . . . . 77
4.3 ATM LAN Emulation . . . . . . . . . . . . . . . . . . . . . . . . 81
4.3.1 LAN Emulation Architecture . . . . . . . . . . . . . . . . 81
4.3.2 Components of ATM LAN Emulation Architecture . . . . 83
4.3.3 Communication In ATM LAN Emulation . . . . . . . . . 84
4.4 Virtual LAN (VLAN) . . . . . . . . . . . . . . . . . . . . . . . . 87
4.5 ATM Reference Model . . . . . . . . . . . . . . . . . . . . . . . . 89
4.6 Interim Local Management Interface(ILMI) . . . . . . . . . . . . 90
4.7 ATM Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
4.7.1 M1 Interface: Management of ATM Network Element . . 92
4.7.2 M2 Interface: Management of Private Network . . . . . . 95
4.7.3 M3 Interface: Customer Network Management . . . . . . 95
4.7.4 M4 Interface: Public Network Management . . . . . . . . 95
4.7.5 M5 Interface: Inter NMS Communication . . . . . . . . . 96

5 Network Management Applications 97

5.1 Configuration Management . . . . . . . . . . . . . . . . . . . . . 98
5.1.1 Network Provisioning . . . . . . . . . . . . . . . . . . . . 100
5.1.2 Inventory Management . . . . . . . . . . . . . . . . . . . 100
5.1.3 Network Topology . . . . . . . . . . . . . . . . . . . . . . 101
5.2 Fault Management . . . . . . . . . . . . . . . . . . . . . . . . . . 101
5.2.1 Fault Detection . . . . . . . . . . . . . . . . . . . . . . . . 102
5.2.2 Fault Location and Isolation . . . . . . . . . . . . . . . . 102
5.3 Security Management . . . . . . . . . . . . . . . . . . . . . . . . 102
5.3.1 Policies and Procedures . . . . . . . . . . . . . . . . . . . 103
5.3.2 Firewall . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103

Telecom Network Management By Nirav Patel

CONTENTS 3

6 Telecommunication Management Network 105

6.1 Difference Between TMN and Communication Networks . . . . . 106
6.2 TMN Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . 107
6.2.1 Functional Architecture . . . . . . . . . . . . . . . . . . . 107
6.2.2 Physical Architecture . . . . . . . . . . . . . . . . . . . . 110
6.2.3 Informational Architecture . . . . . . . . . . . . . . . . . 111
6.3 TMN Logical Level Architecture . . . . . . . . . . . . . . . . . . 112
6.3.1 Element Management Layer . . . . . . . . . . . . . . . . . 113
6.3.2 Network Management Layer . . . . . . . . . . . . . . . . . 114
6.3.3 Service Management Layer . . . . . . . . . . . . . . . . . 114
6.3.4 Business Management layer . . . . . . . . . . . . . . . . . 115
6.4 TMN Implementation . . . . . . . . . . . . . . . . . . . . . . . . 115

Telecom Network Management By Nirav Patel

CONTENTS 4

Telecom Network Management By Nirav Patel

Chapter 1

Overview of Network
Management

5
1.1. INTRODUCTION 6

1.1 Introduction
The phenomenal success of IT and Telecommunication would not have been
possible without any effective management framework. The management tech-
nologies have also been maturing with evolution of IT & Telecom. Besides
having efficient networks a major task for any organization is to manage these
networks efficiently. This involves upkeep of various networking component. In-
ternet which is a network of networks is becoming increasingly complex. It has
become a part of almost every sphere of society, from technology to life style,
from business to politics, its every where but given all these advantages, it has
also resulted in more complex and more heterogeneous network. Different type
of networks makes it even more difficult to have a single management policy
in place to take care of all possible networks. This need gives rise to Telecom
Network Management (TMN).
TMNs provide the means used to transport, store and process information
used to support the management of telecommunication networks and services.
TMNs can be used for the management of telecommunication networks oper-
ated by Administrations, Recognized Operating Agencies (ROAs), customers, or
other organizations and individuals. When these telecommunication networks
are connected with each other, their TMNs provide the means of exchanging
information required to manage end-to-end telecommunication services.
All types of telecommunication networks and network elements such as
analogue networks, digital networks, public networks, private networks, switch-
ing systems, transmission systems, telecommunication software, and logical re-
sources of the network (such as a circuit, path, or telecommunication services
supported by these resources) are candidates for management by a TMN. TMN
can be considered as a network within the telecommunication network responsi-
ble for upkeep, administration and maintenance of the telecommunication net-
work. There is in principle no limit to the fields of application.
The term TMN is introduced by the ITU-T as an abbreviation for ‘Telecom-
munications Management Network’. The concept of a TMN is defined by
Recommendation M.3010. M.3010 is the document specifying the standards
that have to be used by any organization or individual trying to implement
a TMN. TMN has a strong relationship with OSI management, and defines a
number of concepts that have relevance for Internet Management.
According to ITU-T recommendation “TMN is a conceptually separate net-
work that interfaces a telecommunications network at several different points”.
This implies that TMN components will have their own connections with TMN
entities. These entities will be responsible for management of the network, but
they in no way will assist or carry out any task that is related to transmission of
user data. This implies that there will be a clear distinction between TMN and
the Data Network. Figure 1.1 shows the conceptual relation between a TMN
and a data network.
According to the image, TMN interfaces with telecommunication network at
multiple places, this interfacing is used to collect important information about
status of the network. The status is then processed by an operating system,

Telecom Network Management By Nirav Patel

CHAPTER 1. OVERVIEW OF NETWORK MANAGEMENT 7

Figure 1.1: General Relationship Between TMN and Telecommunication Net-

work

which is essentially a program that is operating the application layer of OSI

model. The Operations Systems perform most of the management functions;
these functions may be carried out by human operators but also automatically.
It is possible that a single management function will be performed by multiple
Operations Systems. In this case, the Data Communication Network is used
to exchange management information between the Operation System. Network
management involves activities like Operations, Administration, Maintenance
& Provisioning commonly termed as OAM&P. Traditional telecommunication
networks had a clear distinction between telephone network and data communi-
cation network. With advent of advanced communications system and internet
this separation is becoming blurred. Now a days data networks are made capable
of carrying real time voice communication. This integration though improves
users satisfaction by providing a bundle of services into a single package they
create a whole different set of problems for network managers and administra-
tors. Following few sections will discuss some general cases for a network failure
and the major problems and challenges faced by IT managers administrating a
corporate telecommunication network.

1.1.1 Common Network Problems

Network failures can lead to loss of productivity as important services can be
inaccessible to the users. In case of corporate networks most of the services op-
erate in a client/server mode. Traffic flow is generally bursty and its important
that the network is reliable. Some of the common problems that result into
network failure are

Telecom Network Management By Nirav Patel

1.1. INTRODUCTION 8

• Connectivity Failures: Corporate networks consists of large number

of nodes. These devices include work stations and connecting devices
like hubs, switches and connecting cables (passive devices). All these are
devices that have to be managed by TMN. A breakage of any cable or
failure of any node makes the network inaccessible. Most of the time its a
failure of a connecting node. Also the entire node does not fail at once it is
the connecting interface (port) on the node that fails. This disconnects all
the workstation or nodes connected to that port. One major cause of this
is excess traffic received on that port or interface which will overwhelm
the processing capability of the port.
• Improper IP address allocation: With large organization the role of
network management is delegated to more than one network administra-
tors. IP addresses are logical numbers and they have to be paired with
a physical address (MAC) for proper delivery of messages. mistakes are
made in assigning IP address where two workstations may be assigned
same IP address. This will result into IP address conflict.
• Traffic overload: Bursty nature of communication can result into traffic
overload in certain parts of network. This may overwhelm the network
node that is handelling the traffic an result into its failure.
• Power failures: Power hits could reset network component configura-
tion causing network failure. Network components have some permanent
configurations and some dynamic configurations. A power hit may change
some of these configurations resulting into failure.
With ever increasing size of networks and increased connectivity to internet
(public networks) security of internal corporate network also becomes an im-
portant problem. These matter is more related to policies than to technical
components , but it none the less falls into the scope of network management.

1.1.2 Challenges of IT Managers

Managing a corporate network presents large number of difficulties mainly be-
cause of the size and variation of networks. Corporate networks are becoming
larger and more complex. When the term network management is used, its
scope not only manages the technical aspects of a communication network but
it also includes taking management decisions. Network management not only
deals with networking components but also involves managing and monitoring
flow of traffic through the network. Any breakage in this may result in loss of
service to users. Some of the general problems that an IT manager has to deal
with are
1. Managing Multiple Networks: In client/server communication the
network management part is no longer centralized. The computer and
telecommunication networks are slowly merging and it becomes the re-
ponsibility of IT manager to manage multiple networks. With explosion

Telecom Network Management By Nirav Patel

CHAPTER 1. OVERVIEW OF NETWORK MANAGEMENT 9

of IT services, management of data storage and telecommunication net-

works now falls under the scope of responsibility of a network manager.

2. Security Management: Along with management of network compo-

nents, an IT manager must also control and monitor access to information.
This involves creating policies as to who accesses the information and is
that access made by authentic user. Along with information management,
the corporate network has to be secured against external threats. This is
done using components like firewalls and by putting in place proper data
access policies. Security management in form of authentication and au-
thorization is another aspect of IT manager’s responsibility.

3. Keeping a check on latest technology: Role of an IT manager is also

sometimes synonymous with the term CIO, Chief Information Officer. IT
manager has to keep track of the latest technology and development hap-
pening in the field of communications. Along with updated information
a manager should also have foresight in predicting future trends while se-
lecting any technology for a corporation. This is because his choice will
amount into a huge investment for the corporation and the success or
failure of the decision can cause huge monetary set back.

1.2 Network Management: Goals, Organization

and Functions
Network management functions can be broadly summarized as OAM&P i.e.
Operations Administration Maintenance and Provisioning. These four activities
when performed systematically can be called as efficient network management.

• Operations: Refers to providing IT services to users of corporate net-

work. These service are mostly in client/server mode.

• Administration: This activity is responsible for creating over all goals

policies and procedures for network management. Administration creates
formal documents and processes that need to be practiced for efficient
network management.

• Maintenance: Maintenance activity is mainly the function of Installation

& Maintenance team. Their responsibility involves repair and installation
of facilities and equipment.

• Provisioning: This functions refers network planning and activities which

that help introducing new services in the network.

1.2.1 Goal of Network Management

Goal of network management is to ensure that users of network receive IT
services at an agreed level of QoS. To meet this requirement management should

Telecom Network Management By Nirav Patel

1.2. NETWORK MANAGEMENT: GOALS, ORGANIZATION AND
FUNCTIONS 10

Figure 1.2: Network Management Functional Groups and Activities

establish policy by creating a Service Level Agreement (SLA) for the users.
This SLA will vary depending upon the type of services. SLA is more stringent
for critical services which may need to run uninterrupted 24 x 7. Network
management will not only involve technical aspects but also business aspect
where policy decisions have to be taken with a view of making profit. From a
business administration point of view, network management involves strategic
and tactical planning of engineering, operations and maintenance of network
and network services for current and future needs at a minimum over all cost.
Network management consists of three major groups that take care of overall
network management activities.

1. Network Provisioning

2. Network Operations

3. Network Installation and Maintenance (I&M)

Figure 1.2 shows list of activities performed by individual groups. Although

these activities are assigned to a specific groups, actual tasks pertaining to these
activities are carried out by teams as per companies policies and orgaizations
structure for managing operations. Network Provisioning is primarily related
to planning and design of corporate network. These activities are taken care
by engineering group. Monitoring and administration of activities falls under
Network Operations Centre (NOC) while the actual tasks of maintenance, repair
and installation is carried out by I&M team.

Telecom Network Management By Nirav Patel

CHAPTER 1. OVERVIEW OF NETWORK MANAGEMENT 11

Figure 1.3: Network Management Flowchart

Figure 1.3 shows functional relations between activities of various network

management groups. Engineering group is responsible for policy decisions, for-
mulating business strategies along with planning and design of network manage-
ment. NOC is a core group that performs various task of handling complaints
and tickets raised for any service related issues. All network management groups
rely on other two groups for various information and input. This is represented
in Figure 1.3 where engineering group takes policy decisions based on inputs pro-
vided by NOC and I&M. NOC handles incoming service requests from users and
delegates and maintenance or repair related tasks to I&M. Following sections
elaborate the functions performed by each of network management groups.

1.2.2 Various Functions of Network Management

1. Engineering Group: Network Provisioning
Network Provisioning consists of planning and design and is the core re-
sponsibility of engineering group. As illustrated in Figure 1.2 and Figure
1.3 engineering group is responsible for taking policy and business deci-
sions. This involves keeping track of current technology trends and selec-
tion of appropriate option for designing corporate network. This decisions
are based on thorough analysis of traffic patterns and requirements for a
particular corporate network.
Along with procurement of new equipment efficient use of available re-
sources is also a task assigned to engineering group. This is achieved

Telecom Network Management By Nirav Patel

1.2. NETWORK MANAGEMENT: GOALS, ORGANIZATION AND
FUNCTIONS 12

through inventory management. Another set of responsibility involve cre-

ating policies and procedures for various practices performed by NOC and
I&M. Traffic analysis is done using automated tools. These tools form a
part of Network Management network and help tune the performance of
automatically.

2. Network Operation Centre

Figure 1.3 list various functions performed by NOC. This is the heart of
network management functions where most of the core managing activi-
ties are performed. NOC is concerned with daily activities of network and
providing network services. Some of the functions of NOC are fault man-
agement, configuration management, performance, security and account
management. NOC is responsible for gathering statistics and generating
reports for management. This traffic analysis is done using network man-
agement tools and systems which help automate the system. Following
are the activities of NOC in a detailed fashion.

• Fault Management/ Service Restoration

Service failure or unavailability of any IT service is a common fault
with network management. restoring a service for users is a responsi-
bility of NOC. This involves detection and isolation of problem that
caused service failure. In some situations, network will do this au-
tomatically. Such networks are called self-healing networks. But in
some situations when automatic recovery or healing is not possible,
Network Management Systems (NMS) are used to detect failures.
These failures can be in form of component failure and a NMS raises
appropriates alarms and sends notifications to the concerned team.
When a service failure occurs a ticket is generated either automat-
ically ( NMS has that option) or manually by an engineer. This
ticket contains detailed description about the problem. All the pos-
sible available information about service failure event is present in the
ticket. This ticket is then assigned to I&M group for actual main-
tenance or repair work if necessary. Generally any ticket contains
tracking number of the request, affected service, affected user, any
SLA that is associated with the disrupted service. Once service has
been restored and the underlying problem resolved, ticket is closed.
• Trouble Ticket Administration
Trouble ticket administration is administrative part of network man-
agement system which is used to keep track of all problems in a
network management system. Ticket is a single docket of informa-
tion about a fault that occurs in network. It is kept open until it
is resolved. Generation of trouble ticket is performed automatically
or manually and they are stored in a database. These can later be
reviewed for analysis and for finding pattern or trends in a particular
service fault.

Telecom Network Management By Nirav Patel

CHAPTER 1. OVERVIEW OF NETWORK MANAGEMENT 13

• Configuration Management
Network Configuration Management is the process of organizing
and maintaining information about all components of a computer
network. When a network needs repair, modification, expansion or
upgrading the administrator refers the network configuration man-
agement database. The database contains location and network ad-
dresses of all hardware devices as a well as information about pro-
grams, versions and updates installed in the network components.
There are three types of configurations, ‘static configuration’ (per-
manent configuration), this is the configuration from which the net-
work would start if its started from an idle status. Second configura-
tion is ‘current configuration’ this is the configuration on which
the network is currently running. It is not necessary that current
running configuration is same as static configuration. Third type
of configuration is the ‘planned configuration’, this refers to the
configuration to which the network will be upgraded in future.
The NOC has a display that reflects the dynamic configuration of
network and its status. Any changes in the configuration are reflects
in the NMS tool used by administrators. In case of a service fail-
ure some temporary changes may be added to a components, these
dynamic change are reflected in NMS tool.
• Security Management
Security management can cover a wide range of aspects which in-
cludes physically securing network components as well as securing
access to information to authorized users. a security database is
maintained by NOC for access to the network and network infor-
mation. Other aspects of security management involves maintain-
ing firewalls and using cryptography for encrypting user information
while using public networks.
• Performance Management For optimum performance NOC main-
tains up to date information about the network. Data gathered is
used to tune the network for optimum performance. The network
statistics include data on traffic, network availability and network
delay. Traffic data is made available in various formats where it can
evaluated based on its nature e.g. web traffic, voice traffic, e-mail and
network management traffic. Traffic statistics are helpful in predict-
ing trends and planning future needs. Performance data available on
delay and availability can be used for tuning the network to increase
reliability and improve response time.
• Accounting Management
NOC decides the cost and allocates the use of the network. Metrics
are established to measure the usage of these e-resources and services.

3. Installation and Maintenance (I&M)

Telecom Network Management By Nirav Patel

1.3. NETWORK MANAGEMENT ARCHITECTURE 14

(I&M) takes care of all maintenance and repair work that is carried out
under the process of network management. This group takes orders and
directives from NOC to perform any physical maintenance such as instal-
lation of components and fixing troubles of network operations. Tickets
generated for any service breakdown or maintenance is handed over to
(I&M) for resolution.

1.3 Network Management Architecture

Previous sections have clearly established the need for network management as
well as scope of work under network management. NMS is an important com-
ponent for manging network. Its an application that monitors various network
components. NMS systems provides the user with the dynamic status of net-
work and issues alerts in case a network component goes down. A corporate
network will consist of interconnection of various network components. These
components are not generally from the same vendor. In most scenarios network
will consist of components that are from different vendors running a different
version of operating system.
Each type of component is most efficiently managed by a Network Manage-
ment System developed by its vendor. A NMS allows easy exchange of data
between the application and component making task of network management
simpler. But as mentioned earlier in a corporate network, components from
different vendors have to interact and exchange messages. This might require
need for different management systems for different parts of a corporate net-
work. There is a need for a unified NMS that can be manage devices from
different vendors seamlessly. Since interoperability is such an important con-
straint standards were developed for managing networks. Some of the standard
that are used for management of network are The Internet standard and other
is the OSI standard. More recently there has been a need for automatic network
management systems that have the capability to diagnose a fault and automati-
cally take corrective actions. This has led to new network management systems
like Common Object Request Broker Architecture (CORBA).
ITU Recommendation X.701 specifies three main components of any network
management architecture. These component are shown in Figure 1.4.

1. Functional Components:
This component defines that functions which can be performed by a net-
work management architecture. Five functions have been defined by ITU
recommendation these are fault management, security, accounting, per-
formance management and configuration management.

2. Information Component:
A classical view of network management architecture consists of a network
management system which communicates with an agent which in turn will
communicate with the managed devices. These devices are called managed

Telecom Network Management By Nirav Patel

CHAPTER 1. OVERVIEW OF NETWORK MANAGEMENT 15

Figure 1.4: Domains of Network Management Architecture

Figure 1.5: Network Management Dumbbell Architecture

devices and the NMS creates a view or image of these managed devices
based using information provided by the agent. Information component
specifies the type and structure of information that can be collected to
create a representation of managed devices. This image is called Infor-
mation Model.

3. Communication Component:
This component is responsible for providing reliable communication be-
tween entities of network management architecture. This level defines
various protocol that are used for establishing such reliable communica-
tion.

A network management dumbbell architecture for interoperability is shown

in Figure 1.5. Here components from vendor A and vendor B exchange informa-
tion with each other. This information is generally of two types, ‘management
information data’ which contains device type, id, status and current running
configuration. Other type of information conveyed consists of ‘management
control information’ e.g. change in setting or configuration. Figure 1.5 shows
the functional relation between components from two separate vendors.
Protocols and services associated with dumbbell architecture are presented
in Figure 1.5, Application related services are generally management related

Telecom Network Management By Nirav Patel

1.3. NETWORK MANAGEMENT ARCHITECTURE 16

Figure 1.6: Network Management Components

Figure 1.7: Network Management Interoperability

services like fault configuration management or configuration management ex-

plained in section 1.2.2. Management protocols used for networks are CMIP
Common Management Information Protocol and SNMP Simple Network Man-
agement Protocol. Transport protocols are the first four layers of OSI and
TCIP/IP(first 2 layers) model.
These basic architecture shown in Figure 1.5 and Figure 1.6 can then be
modified to extend the scope and functionality of network management. Figure
1.7 shows a hierarchical model for network management, different agents manage
a set of components which then provide the status report to a NMS. This update
can be periodic as well as based on Polling mechanism or Query sent by the NMS.
The same hierarchical architecture can then extended to have interoperability
between network management systems from different vendors. This architecture
is shown in Figure 1.7 where two NMSs connected to two different network
exchange control and status information.

Telecom Network Management By Nirav Patel

CHAPTER 1. OVERVIEW OF NETWORK MANAGEMENT 17

Summary
This module of Telecom Network Mangement presents an overview of various
terms and terminologies that form the idea of TNM. Apart from technical as-
pects the module also covers management and business related aspects of the
practice of network management. Various qualities necessary to be an IT man-
ager give a brief idea of the skill set necessary for being a successful IT manager.
Also the module summarizes that network management is a collective activity
requiring wide range of practices like research, data analysis carried out by en-
gineering group, maintenance activities planned and proposed by NOC and the
installation and repair work done by I&M team. Next module talks about OSI
management protocol that was developed for network management providing
details about various network management standards and model implement in
actual corporate networks.

Telecom Network Management By Nirav Patel

1.3. NETWORK MANAGEMENT ARCHITECTURE 18

Telecom Network Management By Nirav Patel

Chapter 2

OSI Network Management

19
2.1. INTRODUCTION 20

2.1 Introduction
Network Management has become an integral activity for any corporate net-
work. The previous chapter ‘Overview of Network Management’ intro-
duced general idea about the process of network management. This chapter fo-
cuses of understanding various network management standards developed over
the course of time as well as understand in detail various components of network
management architecture.
Traditionally the management frameworks were designed keeping in view
the demands of specific technology or network. Network management systems
were developed for vendor specific components which created interoperability
issues. Managing heterogeneous network systems is a difficult task because each
of these networks has its own management system. These networks usually are
constructed on independent management protocols which are not compatible
with each other. This results in the coexistence of many management systems
with different managing functions and services across enterprises. Incompati-
bility of different management systems makes management of whole system a
very complex and often complicated job.
The emergence of next generation networks & services has ushered in a
new era of technological advancement. At this time, the focus was to have
some technology-independent and completely autonomous management
framework for networks and its related services. As the need for network man-
agement increased there was a need to standardize the process. Keeping this
view in mind various standards were developed for network management. These
standard tried to address the problem of interoperability between different NMS
and at the same time create a technology independent network management
standard. Having a standard process for network management made it possible
to have distributed management systems.
Two leading models of network management are the Internet Model and OSI
model. Internet model is more widely used because of its simplicity. OSI model
on the other hand is based on object- oriented methodology and is complex in
its implementation. Although complex in implementation OSI based models are
gaining greater importance due to emergence of heterogeneous networks with
components from different vendors. Some of the latest management protocols
and models are based on OSIs object oriented methodology. These standard
are described in greater details in the next sections.

2.2 Network Management Standards

Several network management models have been developed for having a standard
network management activity. Some of these standards are
• OSI Model
• Internet Model
• Telecommunications Management Network (TMN)

Telecom Network Management By Nirav Patel

CHAPTER 2. OSI NETWORK MANAGEMENT 21

Figure 2.1: Network Management Functions Overview

• IEEE LAN/MAN model

• Web based management model
Each network management standard consists of components for network man-
agement, Protocol for management and Information model used for management
process. Each standard defines the type of services that can be performed by
management systems. These service are fault management, configuration man-
agement, security and accounts management and performance management.
General view of any network management system is shown in Figure 2.1.
1. OSI Model :
OSI model was developed by International Standards Organization (ISO),
for developing standarized management process. Protocol structure used
by OSI is called Common Management Information Protocol(CMIP). Re-
gardless of the resource being managed, the protocol defines a basic set
of operations applicable to network management. While CMIP specifies
the protocol, Common Management Information Service (CMIS) specifies
basic services needed to perform various management functions. These
service are shown in Figure 2.1.

Telecom Network Management By Nirav Patel

2.2. NETWORK MANAGEMENT STANDARDS 22

OSI model is based on object oriented philosophy where the management

system will model various managed devices using various attributes. These
attributes are used to create a picture of managed device which will then
be used by the network management system. The collection of such in-
formation is stored in a database called as Management Information Base
(MIB) (more details in section 2.3.1). While modelling individual man-
aged devices OSI model uses concepts of Abstraction 1 and Inheritance 2 .
CMIP/CMIS apart from specifying management protocols and functions
also address network management applications. Both LAN/WAN can be
managed using CMIP/CMIS. Major drawbacks of OSI model are that it is
complex to implement and the protocol stack of CMIP is large and takes
a lot of memory. Even though complex the drawbacks are far insignificant
as compared to the advantages of OSI model they are namely
• Deals with all seven layers
• Most complete set of standards
• Well structure and layered
• Can be used to manage both LAN and WAN
2. Internet Model:
The Internet Engineering Task Force (IETF) is responsible for all specifi-
cations of internet and the related network management standard. Simple
Network Management Protocol (SNMP) is the network management pro-
tocol developed by IETF. These traditional schemes such as SNMP were
more technology specific, network centric, centralized and/or weakly dis-
tributed management schemes. It was widely accepted in industry due to
its simpler design, and architecture.
SNMP architecture is based on Manager-Agent paradigm (more details in
section 2.3.1). It includes five basic components, manager, agent, managed
device, management information and network management protocol. The
manager is the brain of network management and it administrates and
conducts Operations and Management tasks (O&M) for the managed 3
devices. SNMP makes use of various databases like Structure of Man-
agement Information(SMI) and Management Information Base(MIB) for
creating a picture of managed network. SNMP is application layer proto-
col which is used to access managed objects. This management scheme is
named after this protocol. Some of the drawbacks of SNMP protocol are
• SNMP protocol stack with fewer operational commands is inadequate
and could not provide scalability.
1 Abstraction: refers to providing only that information which is necessary for the man-

agement system to get a definite status of managed devices while hiding other critical infor-
mation.
2 Inheritance: allows OSI model to reuse existing components of the Network mangement

architecture.
3 Routers, access servers, switches, hubs and printers etc can be termed as managed devices.

Telecom Network Management By Nirav Patel

CHAPTER 2. OSI NETWORK MANAGEMENT 23

• SNMP is based on connectionless protocol UDP which makes it un-

reliable.
• Inheritance is not used in SNMP and this prevents it from reusing
existing components
• SNMP framework cannot reconfigure managed elements automati-
cally.

3. Telecommunications Management Network:

Telecommunications Management Network(TMN) is developed by Inter-
national Telecommunication Union (ITU) and is based on OSI CMIP/CMIS
specifications. TMN extends the concept of mangement beyond manage-
ment of network. TMN includes a logical architecture that allows the
mangement process to go beyond simple network management. It ad-
dress need for service and business management. Some of the functions
performed by TMN are

• Optimizing network performance.

• QoS management, fault management, accounting, customer coordi-
nation.
• Vendor specific management, Log records, mediation, updating firmware
and fault detection.

4. IEEE Standards for LAN/MAN:

IEEE standard for management of LAN/WAN is concerned with OSI
layer1 (physical) and layer2 (data link). They are structured similar to
OSI specifications. Both OSI/CMIP and Internet/SNMP protocols uses
IEEE standards for management of lower layers.

5. Web Based Management:

Web based network management technologies make use of web servers to
provide network management. These technologies enable remote monitor-
ing and management of network using protocols like XML, HTTP. Web
based network management technologies are based on open standard, pro-
vide interoperable integration across heterogeneous environment.The use
of XML enables distributed applications to behave as loosely coupled while
HTTP(-S) provides not only secure communication but it also enables
communication to reach every nook and corner of the network without
any restriction from firewalls.
Two main technologies that are used for web based management are Web
Based Enterprise Management (WEBM) and Java Management Extension
(JMX). Desktop Management Task Force(DMTF) is developing specifica-
tions for WEBM. DMTF is using an object oriented management model
for developing its standard. JMX makes use of Java applets for manage-
ment. These applets run on network components.

Telecom Network Management By Nirav Patel

2.3. NETWORK MANAGEMENT MODELS 24

2.3 Network Management Models

Before understanding the overview of individual components there is a need for
understanding some of the basic terms used in network management systems.

Terms used in Network Management Models

1. Management Information Base (MIB):
The Management Information Base (MIB) is a conceptual data store that
contains a management view of the device being managed. The conceptual
data contained in this data store constitutes the management information.
MIB should not be confused with a real database. It doesnot contain
actual information about device. It only contains information about the
device which is needed by network management system. NMS creates a
view of various network devices, this view is called the MIB. MIB discusses
the semantics of objects i.e. the information about the object modeled.
2. Structure of Management Information SMI
SMI is responsible for defining the general rules for naming objects (hard-
ware and system, non-physical such as programs, and administrative in-
formation), defining objects types, and show how to encode objects and
values. In case of network management SMI defines the syntax used for
naming objects. Defines how an object should be named while creating a
model.
SMI is a protocol that defines these rules. However, we must understand
that SMI only defines the rules; it does not define how many objects are
managed in an entity or which object uses which type. SMI is a collection
of general rules to name objects and to list their types. The association
of an object with the type is not done by SMI.
3. Manager
Manager refers to the application that performs the task of network mange-
ment. The manager is the brain of network management and it admin-
istrates and conducts Operations and Management tasks (O&M) for the
managed devices.
4. Agent
Agent is software that resides in managed devices and it facilitates the
communication and management tasks between manager and managed
devices. Agent is the mediator between managed device and the manger.
The agent keeps performance information in a database. The manager
has access to the values in the database.

OSI model is one the most structured and well defined model for network
management. All the four components mentioned below are present and well
defined in OSI model for network management. Other standards have a subset of

Telecom Network Management By Nirav Patel

CHAPTER 2. OSI NETWORK MANAGEMENT 25

Figure 2.2: OSI Network Management Model

these four components incorporated into their definition. Network management

model shows various components of a network management standard. Figure
2.2 shows various components of a network management model. It comprises of
4 components namely
• Organization Model
Organization model describes the components of network management
system, their functions and their infrastructure. It defines the terms agent,
object and manager. Different standard have different definitions incorpo-
rated under organization model. SNMP the internet standard has an
organization model similar to OSI model. In contrast IEEE standard for
LAN and WAN deals only with layer 1 and layer 2 and does not have a
well defined structure of organization model.
• Information Model
Information model defines the structure and organization of management
information. Information model specifies the semantics4 and syntax5 of
information used by the agents to perform network management. This
information is stored in form of Management Information Base(MIB) and
Structure of Management Information(SMI). Both these data bases are
conceptual and they only represent the information or an image of the
managed device. The actual information about the managed device is
avaialble with a different database (part of configuration management).
• Communication Model:
The third model in OSI is communication model. This model is responsible
for successful communication between managed and managing devices. Its
role is defining structured protocols for communication requirements. This
component has three subcomponents which are
– Management application process that function in application layer.
4 The type of information to be conveyed
5 The way in which the information has to be conveyed

Telecom Network Management By Nirav Patel

2.4. ORGANIZATION MODEL 26

Figure 2.3: Two Tier Network Management Model

– Layer Management for communication between layers

– Layer Operation for communications within a layer
• Functional Model:
The Functional component defines various activities performed by net-
work management systems. These functions are grouped in five areas and
are shown in Figure 2.1. Functions are configuration, fault, performance,
security and accounting. These are defined as system management func-
tions in OSI. The reason for such a grouping was to facilitate rapid and
consistent progress on each category in individual groups, and not to seg-
regate NMSs for each area. Functions from one area will be influenced by
others, and a system may be implemented with n functions from different
areas to meet the business objectives and market needs.

2.4 Organization Model

Organization model describes the components of network management and their
relationships. Figure 2.3 shows a two-tier model. This model gives a concep-
tual understanding of network management process. In network management
paradigm devices can be classified broadly into two categories, Managed devices
and Managing devices. Managed devices are generally network components
that are used to carry telecommunication and voice data namely hub, switches,
routers and data servers. If devices have the capability to communicate with
managing devices (manager) they are called managed devices else they are sim-
ple unmanaged devices with no communication capability with the management
network.
The managed devices have a process (application) running on them which
facilitates communication with management network, this process is called an
agent(agent can also be a separate device with the sole reponsibility of mon-
itoring managed devices). This agent works as a liaison (mediator) between
management network and managed device. Figure 2.3 shows that the man-
ager(managing device) can query the agent about status of managed device.

Telecom Network Management By Nirav Patel

CHAPTER 2. OSI NETWORK MANAGEMENT 27

Figure 2.4: Three Tier Network Management Model

Also the agent can send some alarm to the manager in case of some failure of
fault with the managed devices. The manager receives management data (sta-
tus/ data other than communication data) and takes necessary action. This
data is stored in Management Database (MDB).
Figure 2.4 shows a three tier management model where in large number of
devices need to be managed by a single manager. This three tier model is an
extension of two tier network management model. Here an intermediate device
acts as a manger/agent and stored status and data related to various managed
device within its scope of monitoring. This data is then sent to a manager
which resides about the intermediate device. This manager can use the data for
statistical analysis.
The above two models represent the interaction of network manager with
local devices and with other managers. In case the managed network is spread
over a geographically wide area for e.g. a corporation with local offices spread
over different cities there is a need for distributed network management pro-
cess. This process involves communication between various Managers as well
as communication between agents running different types of Network Manage-
ment Systems. This scenario is explained by Figure 2.5 that shows a network
management operation model with a MoM(Manager of Managers) and multiple
managers with their agents running different type of NMS.
This model is similar to the extension of Dumbbell architecture seen in
section 1.3. An agent generally runs a process in form of a network management
system. This management system has the capability of communication with
managers.

2.5 Information Model

Information model represents how the data is represented in a network manage-
ment environment. It defines the structure(syntax and semantics) and storage

Telecom Network Management By Nirav Patel

2.5. INFORMATION MODEL 28

Figure 2.5: Network Management Model with MoM

of information needed for network management. Object-oriented principles are

used in defining information models. Abstraction and Inheritance are two such
object oriented principles that help in creating information model.

The representation of objects and information relevant to their

management, forms the management information model.

In the above description two important aspects are taken into consideration.
First is that resources to be managed are referred to as objects. The manage-
ment model will create an object of managed resource. This object contains
all necessary information about the managed resource, which may be used by
agents for management functions. Second part is the relevance of information.
A single managed device will generate a lot of information, out of which only
a small set of data may be needed by an agent for performing mangement fuc-
tions. So a model creates an object corresponding to a managed device with
only the information that is needed by the agent. This process of creating an
object with relevant data is called Information Modeling.
Information of network components is passed between agents and mangers
for various functions. Information model specifies the information base to de-
scribe managed object and their relationships. This relationship is defined us-
ing Structure of Mangement Information which defines the syntax of informa-
tion(the structure to represent information). This along with Management In-
formation Base help create the information model. MIB(see section 2.3) defines
the semantics of information (what exactly does the information convey.)
MIB is used both by agent and manger for exchange of information. The

Telecom Network Management By Nirav Patel

CHAPTER 2. OSI NETWORK MANAGEMENT 29

Figure 2.6: Network Management Model with MIB

MIB associated with agent is called Agent MIB and the MIB associated with a
manager is called Manager MIB. MIBs in both the cases represent a repository
of data describing an object being manged. Information contained in MIBs may
defer based on the type of MIB. A manager MIB contains information about all
the components it manages while an agent MIB contains only local information.
Figure 2.6 shows a network management model where a manager is associ-
ated with a MIB. Along with MIB an additional entity management databas(MDB)
is added to the network model. MDB is a real database containing configuration
information about managed devices. This information is different from infor-
mation stored in MIB. Configuration data refers to actual information about
the status of device which may include details like IP configuration, version
of firmware on device and other network related configuration. This network
related configuration can be used by manager to know the type of managed
device.

2.5.1 Management Information Tree and Base

A collection of managed objects and the properties implemented within a sys-

tem using the schema defined by the information model is referred to as a
management information base (MIB).
In network management it becomes necessary to identify managed devices
with some names. Different standards use different naming methodologies. In
Internet standard a scalar structure is used to identify individual network com-
ponents. OSI standard follows a different naming standard owing to its object
oriented methodology. The naming scheme for identifying the management ob-
jects results in a tree referred to as a management information tree (MIT).

Telecom Network Management By Nirav Patel

2.6. COMMUNICATION MODEL 30

Figure 2.7: Comunication Model in Network Management

2.6 Communication Model

Another dimension of successful network management is Communication Model.
This model defines the structure of interface between managed and managing
device. The goal is to enable successful transfer and interpretation of manage-
ment information. Main requirement of successful communication is reliable
transport mechanisms and suitable session establishment between applications
on either end.
When management information is communicated between manager and agent
and between manager and other manager some protocols and rules have to be
followed. Communication Model takes care of three aspects of this communica-
tion.
• Transport Protocol: The protocols used for sending data over the net-
work. The data refers to management data, transport protocol used for
sending management data are similar to those used for transfer of telecom-
munication data.
• Application Protocol: Devices (Manger and Agent) are both running
instance of applications used for exchanging messages. The application
defines message format of communication.
• Commands and Responses: Almost all the communication happen-
ing in the process of network management is in request/response mode.
Manager requests for certain information or issues certain commands for
performing certain actions on the managed devices. Commands and re-
sponse define the structure of the messages exchanged.
Figure 2.7 shows a communication model, its an overview that represents
various components of network management from a communication model per-
spective.
This communication model shows the type of messages that can be ex-
changed between manager and agent. The operation is in form of client/server

Telecom Network Management By Nirav Patel

CHAPTER 2. OSI NETWORK MANAGEMENT 31

Figure 2.8: Protocol used in Communication Model

mode (request/response mode). Manager can request agent for any specific
information on any managed device under the scope of agent or can issue a
configuration change request where the agent performs certain on any of the
managed devices. Once the action is completed agent acknowledges the request
with a suitable response. In some cases even the agent can initiate a commu-
nication with the manager in form of notification or traps to raise alarm about
any of the managed devices. This represents the commands and responses used
in communication model.
Figure 2.8 shows the communication protocol used to transfer information
between managed objects and managing process, as well as between manage-
ment processes. The OSI uses Common Management Information Protocol
(CMIP) and Common Management Information Services (CMIS). While inter-
net model uses Simple Network Management Protocol(SNMP) for communi-
cation. For transport protocols OSI can use both connection oriented as well
as connectionless transport protocols. Figure 2.8 gives a detailed structure of
protocols used by network management systems at various levels.

2.7 Functional Model

Figure 2.1 shows a generic structure of network management system. Functional
component of network management addresses The functional model component
of the OSI model addresses the user oriented application which are formally
specified in the OSI. These are separately represented in Figure 2.9, all the
functions are categorized in 5 categories. The functions are identified in X.700
for OSI management and ITU Recommendation M.3400 for TMN (for further
reading). Functional model will define the type of services performed by any
network management system. The categories are not exhaustive as per OSI

Telecom Network Management By Nirav Patel

2.7. FUNCTIONAL MODEL 32

Figure 2.9: Functional Model Network Mangement

standard there is another categorization that is classifies services that are com-
mon to all network management process, they are categorized under the section
Common Function Management.

• Configuration Management:
This functional area includes functions that allow a management system to
provision resources and services, monitor and control the state of various
managed devices within the network. It address the functions related to
changing the configuration of various network components. Information
related settings and configurations of managed devices are stored in Man-
agement Database (MDB). This information can be retrived by managers
and can be used to change the configuration of managed devices.
Configuration Management is responsible for setting various parameter in
network components. In case of a failure component management per-
forms the task of informing the team responsible for provisioning that
service. Such failure are recorded and maintained in a database which
can later be reviewed for analysis. This configuration data is stored in
Network Management Systems at the Network Operation Centre (NOC).

• Fault Management:
Fault management includes functions that address alarm surveillance, test-
ing, and fault isolation. Alarm surveillance, as the name implies, allows
reporting alarms with different levels of severity along with the possible
cause of alarm. It also provides a summary of the alarms that are out-
standing, and permits the manager to retrieve the alarm information. As
part of OSI management, generic alarms that may be associated with
various resources are defined.
NMS implements fault management by constantly monitoring the network
for any service or device failures. In case of failure an alert is displayed
in the application. Service has to be restored as soon as possible. This
may involve changing some configuration. In several failure situations
the network could automatically detect the problem and perform trouble
shooting procedures for restoration of service. Such networks are called
self-healing networks.

Telecom Network Management By Nirav Patel

CHAPTER 2. OSI NETWORK MANAGEMENT 33

In case the self healing feature cannot resolve the issue there will be a
ticketing system in place that will issue a trouble ticket, which is carefully
reviewed and followed up until service is not restored. NMS have to ca-
pability of tracking any pending trouble tickets and raise periodic alarms
in case some issues are left un resolved. A record of all such tickets are
mantained in the database which can be used later for analysis of patterns
in which fault occur.

• Performance Management:
The performance management area includes functions to monitor perfor-
mance parameters such as errored seconds, number of bad messages, col-
lecting traffic statistics, and applying control to prevent traffic congestion.
Monitoring and controlling the quality of service is another example of a
performance management function. As part of this area, threshold values
may be assigned for the parameters, and when the threshold is exceeded,
events may be generated to inform the management system.
Up-to-date information of traffic statics should be maintained by NOC
for performance analysis. Along with traffic statistics, any updates in the
configuration changes must be updated with a database in NOC. Perfor-
mance management helps identify trends as well as take policy decisions
related to network management. NMS have the capability to gather in-
formation at all protocol layers. The data gathering feature may also be
used to perform analysis on application specific traffic like e-mail traffic,
web browsing traffic, file transfers and voice calls. Performance data on
delay and availability will help improve response time.

• Security Management:
Security management defines a broad range of activities to be performed
for securing the network. This involves the following aspects

– Physical Security of the network i.e. the network components should

be well guarded against any physical damage.
– Secure access to network resources: Providing access to information
and network resources only to authenticated users. Also making sure
the privilege levels for accessing information are well defined.
– Secured Communication: This feature of communication deals with
securing the transmitted information. Management information along
with telecommunication data should be securely transmitted over the
network. Secured communication prevents tampering of information
as it travels through the network.

Security database is maintained by NOC for access to the network and

network resources. Any unauthorized access will raise an alarm through
NMS. Another mechanism implemented by security management is de-
ploying firewalls to secure corporate network from external threats. Any

Telecom Network Management By Nirav Patel

2.8. ABSTRACT SYNTAX NOTATION ONE (ASN.1) 34

intrusion or attempt to override security procedures are recorded by NMS

systems.

• Accounting Management:
This functional area includes collecting usage data for the resources used
in providing a service and then generating a bill, applying, for example, the
tariff associated with the service. Here again, depending on the service, the
usage information will vary. For example, a phone service often determines
the length of time the connection was used versus a packet service which
collects data on the number of packets sent. A general mechanism that can
be specialized for a specific service is available as part of the OSI systems
management standards. It is to be expected that while collecting the usage
information and reporting on the values is subject to standardization, the
generation of bills and application of tariffs in generating the bills are
considered outside the scope of standardization.

2.8 Abstract Syntax Notation One (ASN.1)

ASN.1 is the acronym for Abstract Syntax Notation One6 , a language for de-
scribing structured information; typically, information intended to be conveyed
across some interface or communication medium. In network management in-
formation is passed between applications and this information traverses through
various layers. There has to be a standard way to represent management infor-
mation efficiently. ASN.1 does the task of representing management information
in a structured fashion, such that applications can decode it easily. This is in
line with the object oriented philosophy of OSI model for network management.
ASN.1 has been standardised internationally by ITU specifications. It is
widely used in the specification of communication protocols. Prior to ASN.1,
information to be conveyed in communication protocols was typically specified
by ascribing meanings to particular bits and bytes in protocol messages. Each
byte of data had to be interpreted for specific information carried in it. ASN.1
gives a distinct advantage to programmers such that complex informations or
messages can be represented using simpler components. ASN.1 aims to make
communication independent of this encoding scheme. For e.g. A construct
Student can be represented as an object with attributes like Name, Age, Reg-
istration Number etc. So the programmer can describe complicated constructs
related to network management using structures that are composed using sim-
pler data types.
ASN.1 makes use of two aspects to facilitate interoperability between systems
and for providing a data definition language independent of any protocol. These
aspects are Abstract Syntax and Transfer Syntax. Abstract syntax refers
to rules for defining and constructing data structures for objects. This is done
using object oriented philosophy and provides definition to the structure of data.
6 The “One” was added to the ASN name by ISO to leave open the future possibility of a

better language for expressing abstract syntaxes.

Telecom Network Management By Nirav Patel

CHAPTER 2. OSI NETWORK MANAGEMENT 35

Transfer syntax on the other hand deals with encoding mechanism. The data
generated using abstract syntax is then encoded using any standard encoding
format defined by transfer syntax. BER (Basic Encoding Rule) is a preferred
format for encoding data bits.
The complier interprets this code and provides a run-time code that will be
understood by the machine.

ASN.1 is a data definition language

2.8.1 Module
The fundamental unit of ASN.1 is the module. The sole purpose of a module is
to name a collection of type definitions and/or value definitions (assignments)
that constitute a data specification. A type definition is used to define and name
a new type by means of a type assignment and a value definition is used
to define and name a specific value, when it is necessary, by means of a value
assignment. The only format constraint on type and/or value assignments in a
module is that each must be on a new line. ASN.1 makes use of formal syntax
to describe objects, as an example a simple definition in ASN.1 is given as

<name>::= <definition>

where name defines the entity and ::= symbol is interpreted as “is defined as”.
As a simple example of a student it can be defined as
neha Student::=
{Name=‘‘Neha P.’’,
Age= ‘‘25’’,
Registration-Number= ‘‘11’’,
}
The above example has a name of student as neha which is an object of
module Student. Attributes like name, age and registration number are added
as per the definitions of module Student. In general any module can be defined
as
<module name> DEFINITIONS::= BEGIN
<name>::=<definition>
END
where name may be any data type included within the module module name.
BEGIN and END are necessary to ascertain the boundary of definition of a
module.

Summary
This chapter provides an introduction to various network management stan-
dards developed by ISO, ITU and other standard making organizations. Net-
work management as an activity has various aspects to it and they are formally
categorized in different models, like Information Model, Organization Model,

Telecom Network Management By Nirav Patel

2.8. ABSTRACT SYNTAX NOTATION ONE (ASN.1) 36

Functional Model and Communication Model. Each model is responsible for a

set of activities and procedures that form a part of network management. Inter-
action between various model is taken care by communication model whereas
defining the structure of information being communicated is the responsibility
of Information Model. Functional model specifies various activities that have to
be performed in network management. Finally ASN.1 defines a formal struc-
tured data definition language that can be used to create information model
and define communication protocols between applications running at manger
and managed devices.
In conclusion OSI network management model follows the process of divid-
ing network management activity among various layers and components, where
each of the component is responsible for a specific activity. Network manage-
ment is performed through interlayer communication, although complex in im-
plementation OSI model gives flexibility to network manager to design network
management systems that are efficient and have high degree of customization.

Telecom Network Management By Nirav Patel

Chapter 3

Internet Management
(SNMP)

37
3.1. INTRODUCTION 38

3.1 Introduction
A large part of being a system administrator is collecting accurate information
about your servers and infrastructure. There are a number of tools and options
for gathering and processing this type of information. Many of them are built
upon a technology called SNMP. SNMP stands for Simple Network Management
Protocol.
SNMP is also referred to as Internet Management. SNMP is a protocol that
is implemented on the application layer of the networking stack. The protocol
was created as a way of gathering information from very different systems in a
consistent manner. Although it can be used in connection to a diverse array
of systems, the method of querying information and the paths to the relevant
information are standardized. SNMP was initially developed only for manag-
ing internet, with advancement its scope of management has extended beyond
internet. Now it aims at managing intranet and telecommunication networks.
SNMP based systems are primarily TCPI/IP based systems, but SNMP can
also be used to monitor non TCP/IP based networks through proxy agents.
SNMP has some very specific roles in network management. It defines the
format of the packet to be sent from a manager to an agent and vice versa.
It also interprets the result and creates statistics (often with the help of other
management software). The packets exchanged contain the object (variable)
names and their status (values). SNMP is responsible for reading and changing
these values.
SNMP based systems are simple to implement and this simplicity makes it
the first choice of network administrators while designing any network manage-
ment systems or applications. Many network components have built-it agents
(applications) that can respond to SNMP messages. This adds to the ease of
setting up network management systems. When a new device capable of com-
municating with SNMP based Network Management Systems (NMS) they are
discovered automatically. SNMP commands allow collection of great deal of
information about any managed devices. This makes it easy to design Network
Management Systems that are self healing. The ease of adding a new device
adds to the popularity of SNMP as a tool for network management.
Initially SNMP protocol was developed as a standardization effort by ITU
in response to ISOs OSI network management stack. OSI standard being the
ultimate standard in network management, SNMP was designed to become
similar to OSI standard but that was not realized. With due course of time
various versions of SNMP were released and the first version came to be named
as SNMPv1. SNMPv2 an SNMPv3 followed this initial standard. Figure 3.1
shows the hierarchy of various protocols used in the TCP/IP stack. SNMP is an
application level protocol that helps NMS from various vendors communicate.
Also it is important to note that SNMP messages and protocol is independent
of the protocols that are implemented in the physical layer. A NMS based
on SNMP will use connectionless UDP in the transport layer while IP in the
network layer for dissemination of information. Data link layer and physical
layer can implement any protocol based on the requirement of the network.

Telecom Network Management By Nirav Patel

CHAPTER 3. INTERNET MANAGEMENT (SNMP) 39

Figure 3.1: Protocol Hierarchy in SNMP based NMS

This chapter provides information about SNMP as a network management

protocol. It provides an overview of SNMP model which has a lot of similarity to
network management systems in OSI. The terms used in SNMP based network
management are explained next. Although they have been explained in the
previous chapters their definitions are repeated for ease of understanding.
Organization model of SNMP gives an overview of network management sys-
tems and shows how various components of a management network interact with
each other. System overview of SNMP gives details of how a manger and agent
communicate giving the list of commands and responses exchanged between
them. Information model explains how SNMP creates objects for managing de-
vices and how these objects are stored in MIB. Communication model focuses on
various aspects of communicating management information some of which are
security, having a common communication protocol between NMS from differ-
ent vendors and what view of MIB is presented to what component of network
management. Finally the chapter concludes with a brief description of SNMP
proxy server, the functional model and Remote Monitoring protocol(RMON).

3.2 System Overview : The SNMP Model

Before understanding the SNMP model and the organizational model of SNMP
it is important to have a clear understanding of the various components of
network management systems. These components are

• SNMP Manager:
An SNMP manager is a computer that is configured to poll SNMP agent
for information. The manager can be any machine that can send query
requests to SNMP agents with the correct credentials. Sometimes, this
is implemented as part of a monitoring suite, while other times this is
an administrator using some simple utilities (commands) to craft a quick
request.
Functions performed by a Manger are

1. Queries agents
2. Gets responses from agents
3. Sets variables in agents

Telecom Network Management By Nirav Patel

3.2. SYSTEM OVERVIEW : THE SNMP MODEL 40

4. Acknowledges asynchronous events from agents

Apart from these functions a manger(SNMP manager) also interacts with

another manager agents to give the NOC holistic view of the managed
network.

• SNMP Agents:
SNMP agents do the bulk of the work. They are responsible for gathering
information about the local system and storing them in a format that
can be queried.updating a database called the “management information
base”, or MIB. The agent computer configures which managers should
have access to its information. It can also act as an intermediary to report
information on devices it can connect to that are not configured for SNMP
traffic. This provides a lot of flexibility in getting your components online
and SNMP accessible. Some important functions performed by SNMP
agents are

1. Collects management information about its local environment

2. Stores and retrieves management information as defined in the MIB.
3. Signals an event to the manager.
4. Acts as a proxy for some non–SNMP manageable network node.

• Network Management Sytems:

Network management system are the application or the user interface that
provide real time status of the network along with the past statistics. A
device vendor provides a NMS with the device, this NMS has the capability
to provide large amount of detailed information about any device. It is also
common to have NMS from different vendors interacting with each other
to provide a unified view of the manged network to the administrators.
Some of the advantages of having a network management systems are

1. Consistency in administration
2. Standardized process in managing networks
3. Automated management in case of self healing networks
4. Centralized control provided to administrators in the NOC

Figure 3.2 shows the hierarchy present in a management network.

SNMP protocol operates at the application layer and there is a standard form
in which communication happens between an agent and manager. Manager as a
device is a computer that has the capability to query as well as configure devices
remotely. These functions are carried out using commands between agent and
manager. The data path from application layer to the physical medium passes
through transport layer, internet layer, data link layer and physical layer. The
upper three layers of OSI model are incorporated in SNMP protocol. The
lower layers can implement any communication protocol but usually the set of

Telecom Network Management By Nirav Patel

CHAPTER 3. INTERNET MANAGEMENT (SNMP) 41

Figure 3.2: Hierarchy of Devices in SNMP

Telecom Network Management By Nirav Patel

3.2. SYSTEM OVERVIEW : THE SNMP MODEL 42

Figure 3.3: SNMP Mangement Architecture and Protocol Stack

protocols shown in figure 3.3 are used. The protocol at data link layer and
physical layer change depending upon the transmission medium used. SNMP
mainly uses UDP (connectionless) for transmitting commands. This is one
major drawback as connectionless protocol affects reliability1 of communication.
Figure 3.3 also shows the list of all commands that are exchanged between
SNMP manager and agents. Most important aspect of SNMP is its simplicity is
shown in the fact that there are only 5 protocol messages exchanged between
manager and agent. Out of these 5 protocol messages 3 are initiated by the
manger and the remaining two are initiated by the agents. These protocol
messages are

• get-request: A Get message is sent by a manager to an agent to request

the value of a specific OID (Object Identifier). This request is answered
with a Response message that is sent back to the manager with the data.
The data sent back is a scalar value. This protocol message will help the
manager to query any managed device and obtain certain device specific
parameters.

• get-next-request: A get-next-request message allows a manager to

request the next sequential object in the MIB. This is a way that you
can traverse the structure of the MIB without worrying about what OIDs
to query. This request is used in case multiple values are present for a
particular object. An example of this is routing tables of router which
will have multiple values.
1 With advent of better communication technologies errors in communication are reduced.

Also later versions of SNMP v1 and v2 try to over come this drawback

Telecom Network Management By Nirav Patel

CHAPTER 3. INTERNET MANAGEMENT (SNMP) 43

• set-request: A Set message is sent by a manager to an agent in order

to change the value held by a variable on the agent. This can be used to
control configuration information or otherwise modify the state of remote
hosts. This is the only write operation defined by the protocol. Using
set-request the settable configuration parameter can be changed.

• get-response: This message, sent by an agent, is used to send any re-

quested information back to the manager. It serves as both a transport
for the data requested, as well as an acknowledgement of receipt of the
request. If the requested data cannot be returned, the response contains
error fields that can be set with further information. A response message
must be returned for any of the above requests, as well as Inform messages.

• trap: A trap message is generally sent by an agent to a manager. Traps

are asynchronous notifications in that they are unsolicited by the manager
receiving them. They are mainly used by agents to inform managers of
events that are happening on their managed devices. An example of trap
is when an interface on a router or switch fails the agent monitoring the
device has to send and alarm to the manager, these messages are in form
of trap.

In a later improvement of SNMP another manager initiated message was intro-

duced as GetBulkRequest.

• GetBulkRequest:
The GetBulkRequest PDU is sent from the manager to the agent to re-
trieve a large amount of data. It can be used instead of multiple Ge-
tRequest and GetNextRequest PDUs. So when the manager is aware of
that it has to retrieve an entire table from the agent instead of sending
sequential GetReuest, manager simply makes use of GetBulkRequest.

3.3 Organization Model of SNMP

SNMP models generally operate in client server or more specifically manager/agent
mode. Manager can query and change configuration of managed devices by send-
ing suitable protocol messages to agent of managed device. The organization
model of SNMP tells us how a manger and an agent interact with each other
and what is the hierarchy in which managers and agents are organized in a
management network.
Initial organization of SNMP is a simple two tier model where agent is a pro-
cess running on the manged device and managers is a process (program) running
in the NMS. This organization is shown in Figure 3.4 where both manager and
agent are software modules interacting with each other.
Figure 3.5 shows a two tier organization model where a single manger com-
municates with single agent. This model though simple may not be practical
and model as more than one managers would want to communicate with the

Telecom Network Management By Nirav Patel

3.3. ORGANIZATION MODEL OF SNMP 44

Figure 3.4: Two-Tier Organization Model

Figure 3.5: Two Tier Organization Model with Two Managers

same managed device. This organization model is shown in Figure 3.5 where
multiple managers(mostly from different vendors) communicate with a single
agent.
NMS are generally developed for vendor specific devices, although there are
NMS that can collect information from devices from other vendors the scope of
information is limited. In case of vendor specific NMS the detail of information
collected can be to a greater depth and this helps in faster resolution of faults.
Thus it is common practice to use an NMS to monitor network of multiple
vendor product and several vendor’s NMS to respective network elements.
In the two tier models described above the managers receive raw unprocessed
data, in many cases the managers may want processed data directly from the
agent and use it for taking decisions or for fault finding. The manager cannot
continuously probe the managed device for this data. To avoid this intermediate
devices are inserted that probe the managed device and their agents, collect data
(unprocessed), process the data from various agents and send processed data to

Telecom Network Management By Nirav Patel

CHAPTER 3. INTERNET MANAGEMENT (SNMP) 45

Figure 3.6: Three Tier Organization Model with RMON

Figure 3.7: Organization Model with Proxy Server

manager. Such intermediate devices are agents that perform the task of data
processing and are called Remote Monitoring Agents (RMON). This creates a
three tier architecture and is shown in Figure 3.6.
In practical managed network all the devices are not managed. SNMP man-
agers cannot manage devices that do not have an SNMP agent running on
them. These applications may be legacy telephone systems and wireless de-
vices with no agent running on them. Such devices are managed using PROXY
SERVERS. These servers collect the information in a central fashion and com-
municate with SNMP managers using suitable protocol messages. Figure 3.7
shows such a managed network with a proxy server.
A SNMP management system can behave as an agent as well as a manager.
This is similar to a client/server architecture, where a host may behave like a
client for one part of communication and behave like an agent for the some other
communication. In the two tier architecture shown in Figure 3.6 RMON proble

Telecom Network Management By Nirav Patel

3.4. COMMUNICATION MODEL 46

or agent collects data from various managed devices. Here RMON functions as
a manager but the same device while communicating with a manger will func-
tion like an agent. This shows that both the agent and the manager processes
(programs) are running on mangers. Another case of dual nature of managers
is when NMS from two different vendors communicate to create a holistic pic-
ture of manged network. In this case the NMSs will act like both manager and
agent. So organization model of SNMP shows the following features on a SNMP
managed network
1. SNMP model can be two-tier or three tier.
2. Intermediate processing of data can be done by devices called RMON
probes.
3. Non Managed devices and application than cannot run SNMP processes
on them can be managed by using proxy servers.
4. Manager has dual roles. It can function as manger as well as an agent
based on the type of communication.

3.4 Communication Model

The SNMP communication model defines four important aspects of SNMP com-
munication.
1. How to communicate with other entities? – Architecture
2. Who can communicate what information? – Administrative Policy
3. What protocol is used for communication? – SNMP protocol
4. Which information is communicated? – SNMP MIB (Management In-
formation Base)

3.4.1 SNMP Architecture

SNMP architecture consists of managers and agents, these devices have SNMP
protocol stack2 installed on them. SNMP communication protocol is used to
communicate information between agent and manager. Agents function the task
of aggregating data and supply information (processed or unprocessed) to the
manager. There are three main goals of SNMP architecture
1. Simplicity: Reduce the number and complexity of management functions
performed by the manger.
2. Flexiblity: Architecture should be flexible enough to allow expansion
of manged network and include new managed devices as and when the
network expands.
2 Programms and applications that help the device communicate using SNMP

Telecom Network Management By Nirav Patel

CHAPTER 3. INTERNET MANAGEMENT (SNMP) 47

3. Independent: SNMP architecture should be independent of the archi-

tecture of particular hosts and devices.

First goal of SNMP architecture is realized by having a limited set of com-

mands between manager and manged devices. There are only five protocol com-
mands; three of which are initiated by manager(get-request, get-next-request,
set-request) and two are agent initiated (set-response, trap).
Trap messages are agent initiated and are unsolicited3 message. To keep
SNMP architecture simple the number of unsolicited messages are limited. Trap
are classified into three categories

1. Generic traps

2. Specific traps

3. Time stamp

Each of the trap types are further classified into subtypes. Generic traps
have following types

• coldStart

• warmStart

• linkUp

• linkDown

• authenticationFailure

• egpNeighborLoss

• enterpriseSpecific

Each of these traps is responsible for raising alarm for specific types of event.
Specific traps are used in case enterprise specific traps are not available. Time-
stamp trap is used to find the time elapsed between the initialization (turning
on) and generation of trap for a particular device.
In view to keep communication simple SNMP architecture makes use of
connectionless UDP protocol at the transport layer. It is important to realize
that SNMP is an application layer protocol, it is made such that is independent
of the protocols implemented in the lower layers. UDP is a connectionless
protocol which helps reduce complexity as well as the amount of data that is
communicated. Although UDP is used at the transport layer SNMP is a flexible
protocol and can be used with a wide range of lower layer protocols.
3 alerts and message generated when certain event occurs

Telecom Network Management By Nirav Patel

3.4. COMMUNICATION MODEL 48

Figure 3.8: SNMP Community with Multiple Manager/Agent Pair

3.4.2 Administration Model of SNMP

Communication model aims to maintain simplicity in implementing SNMP pro-
tocol and it does that by reducing the number of communication messages.
Administration Model on the other hand is responsible for secure communica-
tion. The information communicated over the network between managers and
agents should only be visible to authenticated and legitimate devices. This
policy of what data will be visible to which device is implemented using the
administration model of SNMP.

SNMP administrative model helps to understand relationship

between entities that participate in the communication protocol in
SNMP.

Some terms that are used in understanding administrative model of SNMP are

• Protocol Entity: A protocol entity is a device in the SNMP architec-

ture that has SNMP application installed on it and it takes part in the
communication process.

• Application Entity: An application entity is the process of program

installed on a device which makes it possible for the device to communicate
using SNMP protocol.

• Community: The pair of two entities is called SNMP community. The

name given to such a pair is called Community. a community can con-
tain multiple pairs, Figure 3.8 shows a community with three managers
communicating with a single agent.

• Community Profile: Combination of community name,MIB view and

specifications of access privileges. (covered later)

Different pairs of agents/managers become a part of same community by

having a common community name. The authentication process shown in Figure
3.8 is generally done using same community name which is the simplest form of
authentication. This scheme of authentication acts as a filter where agents or

Telecom Network Management By Nirav Patel

CHAPTER 3. INTERNET MANAGEMENT (SNMP) 49

Figure 3.9: SNMP Community

managers (protocol entities) from different community do not receive messages

intended for a different community.
Information about a managed is communicated through objects created for
those device. All the information that is necessary for management tasks are
stored in MIB. A single network element consists of various objects. These
objects can be standard as well as private (visible only to few). A management
agent or manager may not have access to full information about the object i.e.
a device is permitted only a subset of network elements managed objects. This
subset is called MIB view. This means a manager may not have access to all
the information about a device (only that necessary for management is made
visible to a manager).
A MIB view specifies read and write permission for a particular device.
Figure 3.9 shows a scenario where a single agent has different READ/WRITE
permission for 3 objects. This permission for READ or WRITE is called access-
mode. So a community profile can then be summarized as

Community Profile = Community + MIB view + Access-mode

Figure 3.9 shows different access modes, there are four access modes shown
in the figure. These can be mapped to the communication protocol messages
that are exchanged between manager, agent and device.

• Read-Only Access Mode: This mode enables agents to monitor data from
a particular device. Most of the standard information about a device is
always read only. In this case the operations that are possible are get,
response and trap.

• Write Mode: In this mode the manager can send set requests to the device
and change the configuration of a device.

The next term used in SNMP administration model is SNMP Access Pol-
icy. SNMP MIB view when combined with access mode will result in SNMP

Telecom Network Management By Nirav Patel

3.4. COMMUNICATION MODEL 50

Figure 3.10: SNMP Access Policy Single Community

access policy. This access policy gives a complete picture about the views avail-
able to protocol entities and their privileges. All the terms put together (Proto-
col Entity, Application Entity, Community, MIB View Access mode and Access
Policy) define the administration model of SNMP. Figure 3.10 shows a visual
description of SNMP Access Policy with a single community and two community
profiles.
Figure 3.11 shows a complete view of access policy where there are three
communities sharing devices, managers and agents. Different agents have dif-
ferent view (community profiles) and that combined with the community view
gives the SNMP access policy. As described in earlier sections a single man-
ager will have ability to communicate with different agents as well as different
mangers. This is shown in the Figure 3.11 where manager 3 can communicate
with devices and agents from community1 and community 2.

3.4.3 SNMP Protocol Specifications

SNMP protocol specification describes how a SNMP Protocol Data Unit4 (PDU)
is created. Each device runs an application that allows it to communicate with
other devices and pass management information. A more technical description
of this program or application is given in terms of Protocol Entities.
Each device that participates in network management using SNMP runs a
piece of software, generically called an SNMP entity. The SNMP entity is re-
sponsible for implementing all of the various functions of the SNMP protocol.
Each entity consists of two primary software components. The software compo-
nents on a device depends of course on whether the device is a managed node
or a network management station. The peer processes which implement the
SNMP, and thus support the SNMP application entities are called Protocol
Entities.
4 PDU is the message generated by an agent of manager for transfer of information. This

messages is structured in a definite format and passed along the protocol stack where each
layer will add its corresponding header and trailer.

Telecom Network Management By Nirav Patel

CHAPTER 3. INTERNET MANAGEMENT (SNMP) 51

Figure 3.11: SNMP Administration Model

Telecom Network Management By Nirav Patel

3.4. COMMUNICATION MODEL 52

Figure 3.12: Encapsulation of SNMP PDU

All the messages passed by a SNMP device are formatted in a PDU. A

SNMP message is shown in Figure 3.12, where the application layer generates
a PDU encapsulates (surrounds) it with header information and passes it down
along the protocol stack. Header information added to the SNMP PDU contains
Version, Community name along with the application header.
Trasnport layer, network layer and data link layer add their correspond-
ing header and finally the PDU is encoded using a suitable encoding scheme
and passed along the physical medium. A protocol entity receives messages at
UDP port 161 on the host with which it is associated for all messages except
for those which report traps (i.e., all messages except those which contain the
Trap-PDU). Messages which report traps should be received on UDP port 162
for further processing. An implementation of this protocol need not accept mes-
sages whose length exceeds 484 bytes. Since there are only 5 basic types of pro-
tocol messages it is mandatory that all implementations of the SNMP support
the five PDUs: GetRequest-PDU, GetNextRequest-PDU, GetResponse-PDU,
SetRequest-PDU, and Trap-PDU.
Before understanding the construction of a SNMP PDU and the steps in-
volved in communication between source and receiver, there is a need to under-
stand some common constructs that are used in constructing a PDU. the term
transport address is used for specifying source address in the SNMP PDU. In
the case of the UDP5 , a transport address consists of an IP address along with
a UDP port.

Transport Address = Port Number + IP Address

Figure 3.13 shows a basic SNMP PDU that is used to construct a message.
Each term involved in the PDU is constructed using a ASN.1 definition. This
definition specifies the message that is carried by the block within the PDU.

• RequestID:
5 incase some other protocols are implemented at the lower layer, the definition of transport

address should be changed accordingly

Telecom Network Management By Nirav Patel

CHAPTER 3. INTERNET MANAGEMENT (SNMP) 53

Figure 3.13: SNMP Get/Set Type PDU

ASN.1 Definition
RequestID ::=
INTEGER
RequestIDs are used to distinguish among outstanding requests. By use
of the RequestID, an SNMP application entity can correlate incoming
responses with outstanding requests. In cases where an unreliable data-
gram service is being used, the RequestID also provides a simple means
of identifying messages duplicated by the network.

• ErrorStatus:
ASN.1 Definition
ErrorStatus ::=
INTEGER {
noError(0),
tooBig(1),
noSuchName(2),
badValue(3),
readOnly(4)
genErr(5)
}
A non-zero instance of ErrorStatus is used to indicate that an exception
occurred while processing a request.

• ErrorIndex:
ASN.1 Definition
ErrorIndex ::=
INTEGER
In case of a non-zero ErrorStatus, ErrorIndex may provide additional in-
formation by indicating which variable in a list caused the exception.

• VarBind:
ASN.1 Definition
VarBind ::=
SEQUENCE {
name,
ObjectName,
value

Telecom Network Management By Nirav Patel

3.4. COMMUNICATION MODEL 54

ObjectSyntax
}
A managed device is associated with a manged object, this object is termed
as a variable.(different managed objects will have different values). A vari-
able binding, or VarBind, refers to the pairing of the name of a variable
to the variable’s value. A VarBindList is a simple list of variable names
and corresponding values.

3.4.4 Flow of SNMP PDU

The top-level actions of a protocol entity which generates a message are as
follows:

1. It first constructs the appropriate PDU, e.g., the GetRequest-PDU, as an

ASN.1 object.

2. It then passes this ASN.1 object along with a community name its source
transport address and the destination transport address, to the service
which implements the desired authentication scheme.

3. This authentication service returns another ASN.1 object.

4. The protocol entity then constructs an ASN.1 Message object, using the
community name and the resulting ASN.1 object.

5. This new ASN.1 object is then serialized, using the basic encoding rules of
ASN.1, and then sent using a transport service to the peer protocol entity.

Similarly, the top-level actions of a protocol entity which receives a message

are as follows (process followed at the receiver):

1. It performs a rudimentary parse of the incoming datagram to build an

ASN.1 object corresponding to an ASN.1 Message object. If the parse
fails, it discards the datagram and performs no further actions.

2. It then verifies the version number of the SNMP message. If there is a

mismatch, it discards the datagram and performs no further actions.

3. The protocol entity then passes the community name and user data found
in the ASN.1 Message object, along with the datagram’s source and des-
tination transport addresses to the service which implements the desired
authentication scheme (security measures).

4. This entity returns another ASN.1 object, or signals an authentication

failure. In the latter case, the protocol entity notes this failure, (possi-
bly) generates a trap, and discards the datagram and performs no further
actions.

Telecom Network Management By Nirav Patel

CHAPTER 3. INTERNET MANAGEMENT (SNMP) 55

Figure 3.14: SNMP Proxy Server and Firewall

5. The protocol entity then performs a rudimentary parse on the ASN.1

object returned from the authentication service to build an ASN.1 object
corresponding to an ASN.1 PDUs object.

6. If the parse fails, it discards the datagram and performs no further actions.
Otherwise, using the named SNMP community, the appropriate profile is
selected, and the PDU is processed accordingly.

3.5 SNMP Proxy Server

SNMP managers and agents can manage devices that have SNMP processes
installed on them or if they have the capability of handling SNMP messages. In
case of devices that are not SNMP capable a proxy server is used that is used
to convert Non-SNMP entities into SNMP-entities defined by the MIB. Proxy
servers are also used to maintain backward compatibility between two different
versions of SNMP. As is the case with SNMPv1 and SNMPv2, where version 2 is
not compatible with version 1, to make communication possible a proxy server
is used. Finally proxy servers are used in networks that make use of firewalls.
The goal of SNMP proxy is to enable a management application to be able
to perform SNMP operations (e.g. SNMP GetRequest, SNMP SetRequest), on
SNMP agents residing on managed DMZ devices (e.g. routers, switches, servers,
load balancers). SNMP proxy is required whenever the flow of SNMP traffic
is restricted or prevented, due to a firewall installed between the management
application and the managed devices.
Figure 3.14 shows the use of proxy servers in case SNMP messages have to
by pass a firewall. The application-level proxy transports SNMP protocol data
units (PDU) between the NMS and managed devices, allowing requests and
responses between the NMS and managed devices and forwarding autonomous
messages to the NMS. The proxy agent requires little provisioning at the NOC
and no additional provisioning at the managed devices.
The second function performed by SNMP proxy server is where it makes it
possible for SNMPv2 to be backward compatible with SNMPv1. Figure 3.15
shows how SNMP proxy server allows communication between SNMPv1 and
SNMPv2. Request and responses from version 1 and 2 pass through the firewall

Telecom Network Management By Nirav Patel

3.6. REMOTE MONITORING 56

Figure 3.15: SNMP Proxy Server SNMPv1-v2

with some modifications. These modifications are introduced in the message

PDUs by proxy server to maintain backward compatibility. For instance Ge-
tRequest and GetNextRequest pass through the proxy server without modifi-
cations where as, GetBulkRequest is modified by the proxy server by changing
the PDU to make it similar to GetNextRequest. Similar modifications are made
to getResponse and Trap messages.

3.6 Remote Monitoring

SNMP provides a great flexibility in managing network, through NMS a manager
at the NOC can remotely monitor a network. This monitoring though is of
reactive in nature where fault and service management happens when an alarm
or alert is raised by the NMS. An efficient network management process takes
into account the statistical data of the network being managed. By doing this it
can pro-actively avert some situations and solve issues even before they become
critical.
Remote Monitoring refers to a management process where statistical data is
analyzed by devices and timely actions are taken. Remote monitoring involves
using tools that monitor network traffic, they analyze each packet that flows
through the network. The packet flow is not altered the analysis is a passive
process. This analysis is done using actual devices that are a part of the net-
work and tools (applications) that help analyze the raw data that is collected.
The device along with the tools is called a probe. This probe collects network
statistics and the processor analyses data. Once the probes poll information and
data is analyzed locally it can be transmitted to a remote network management

Telecom Network Management By Nirav Patel

CHAPTER 3. INTERNET MANAGEMENT (SNMP) 57

station. In such cases monitoring of a network with a probe is called Remote

Monitoring (RMON).
Remote network monitoring devices, often called monitors or probes, are
instruments that exist for the purpose of managing a network. Often these
remote probes are stand-alone devices and devote significant internal resources
for the sole purpose of managing a network. An organization may employ many
of these devices, one per network segment, to manage it. In addition, these
devices may be used for a network management service provider to access a
client network, often geographically remote.
The next section will describe objects and groups that are used by RMON
to facilitate communication between RMON probes and RMON management
applications. These objects are not manipulated by engineers, their state is
changed by network management applications to indicate the state of network
being managed. RMON1 was specifically designed for Ethernet. Most of the
objects that are a part of RMON1 MIB are used to perform functions specific to
Ethernet. Although their capabilities were later extended to be made suitable
for Token Rings, RMON1 performs function upto the data link layer .

3.6.1 RMON Advantages

• Offline Operation
There are sometimes conditions when a management station will not be
in constant contact with its remote monitoring devices. This is sometimes
by design in an attempt to lower communications costs (especially when
communicating over a WAN or dialup link), or by accident as network
failures affect the communications between the management station and
the probe.
For this reason, this MIB allows a probe to be configured to perform
diagnostics and to collect statistics continuously, even when communica-
tion with the management station may not be possible or efficient. The
probe may then attempt to notify the management station when an ex-
ceptional condition occurs. Thus, even in circumstances where commu-
nication between management station and probe is not continuous, fault,
performance, and configuration information may be continuously accu-
mulated and communicated to the management station conveniently and
efficiently.

• Proactive Monitoring
Given the resources available on the monitor, it is potentially helpful for
it continuously to run diagnostics and to log network performance. The
monitor is always available at the onset of any failure. It can notify the
management station of the failure and can store historical statistical infor-
mation about the failure. This historical information can be played back
by the management station in an attempt to perform further diagnosis
into the cause of the problem.

Telecom Network Management By Nirav Patel

3.6. REMOTE MONITORING 58

• Problem Detection and Reporting

The monitor can be configured to recognize conditions, most notably error
conditions, and continuously to check for them. When one of these condi-
tions occurs, the event may be logged, and management stations may be
notified in a number of ways.

• Exclusive Devices for Monitoring

Because a remote monitoring device represents a network resource ded-
icated exclusively to network management functions, and because it is
located directly on the monitored portion of the network, the remote net-
work monitoring device has the opportunity to add significant value to the
data it collects. For instance, by highlighting those hosts on the network
that generate the most traffic or errors, the probe can give the management
station precisely the information it needs to solve a class of problems.

• Multiple Managers
An organization may have multiple management stations for different
units of the organization, for different functions (e.g. engineering and
operations), and in an attempt to provide disaster recovery. Because en-
vironments with multiple management stations are common, the remote
network monitoring device has to deal with more than own management
station, potentially using its resources concurrently.

3.6.2 RMON Structure of MIB

In remote monitoring MIB consists of 10 groups, each group contains multiple
objects that are used to implement various functions of network management.
These groups are

• statistics

• control

• history

• alarm

• host

• hostTopN

• matrix

• filter

• packet capture

• event

Telecom Network Management By Nirav Patel

CHAPTER 3. INTERNET MANAGEMENT (SNMP) 59

These group forms the basic unit of RMON MIB. They are used to implement
different functions. When a group is used to implement any particular functions
all the objects within that group have to be used. The use of each of these groups
is

• Statistics Group
The statistics group contains statistics measured by the probe for each
monitored Ethernet interface on this device. This group consists of the
etherStatsTable. In the future other groups will be defined for other media
types including Token Ring and FDDI. These groups should follow the
same model as the ethernet statistics group.

• Control Group
The history control group controls the periodic statistical sampling of
data from various types of networks. This group consists of the histo-
ryControlTable.

• History Group
The ethernet history group records periodic statistical samples from an
ethernet network and stores them for later retrieval. This group consists
of the etherHistoryTable. In the future, other groups will be defined for
other media types including Token Ring and FDDI.

• Alarm Group
The alarm group periodically takes statistical samples from variables in
the probe and compares them to previously configured thresholds. If the
monitored variable crosses a threshold, an event is generated.

• Host Group
The host group contains statistics associated with each host discovered
on the network. This group discovers hosts on the network by keeping
a list of source and destination MAC Addresses seen in good packets
promiscuously received from the network.

• HostTopN Group
The hostTopN group is used to prepare reports that describe the hosts
that top a list ordered by one of their statistics. The available statis-
tics are samples of one of their base statistics over an interval specified
by the management station. Thus, these statistics are rate based. The
management station also selects how many such hosts are reported.

• Matrix Group
The matrix group stores statistics for conversations between sets of two
addresses. As the device detects a new conversation, it creates a new entry
in its tables.

Telecom Network Management By Nirav Patel

3.7. RMON FUNCTIONS 60

Figure 3.16: RMON Functions

• Filter Group
The filter group allows packets to be matched by a filter equation. These
matched packets form a data stream that may be captured or may generate
events.

• Packet Capture Group

The Packet Capture group allows packets to be captured after they flow
through a channel.

• Event Group
The event group controls the generation and notification of events from
this device. This group consists of the eventTable and the logTable.

3.7 RMON Functions

RMON1 was initially developed for monitoring Ethernet. Some modification in
the SMI/MIB were made to add capability to monitor Token Ring networks.
10 groups of RMON collectively perform 3 basic functions, these function are
shown in Figure 3.16. It is important to understand that RMON was initially
developed to perform functions related up-to layer 2 of OSI reference layer (Data
link layer).
Three broad categories of RMON functions are

• Statistics Generation

Telecom Network Management By Nirav Patel

CHAPTER 3. INTERNET MANAGEMENT (SNMP) 61

Figure 3.17: Components of SNMP

• Filtering Packets
• Alarm and Alerts Generation
The function of statistics generation is used to provides varied types of statis-
tics, this includes statistics for the ethernet and token ring network being moni-
tored. This also includes providing statistics of data that flows between various
hosts on the network. All these functions are depicted in Figure 3.16.
Ethernet statistics block is the group of statistical data associated with eth-
ernet network. This includes the history block that RMON objets will use for
providing proactive monitoring it contains data gathered from various networks.
Similar analysis is done for token ring networks that are a part of managed net-
work. The first 9 groups mentioned in the list of MIB groups will be used for
providing statistics and management of ethernet networks. The block related to
Host and conversation statistics will maintain data statistics of data transferred
between various hosts. RMON will make use of objects from Host-Top-N group
and Matrix group to provide this data.
Packet filtering group performs the task of monitoring every packet to de-
termine it nature and type. This is done using a set of complicated filters6 .
Finally the function of alarm generation and event logging is provided by the
alarm generation block. For this objects from the group of alarm is used.

3.8 SNMP Management Components

This section gives details about various components of SNMP protocol and how
the definition of SMI and MIB is created for a SNMP protocol.In this section
SMI and MIB are again described in relation to SNMP. Figure 3.17 gives the
relation betweeen various components of SNMP.
Management protocols developed using SNMP protocol include a definition
of objects in form of SMI whereas the process of creating and initializing objects
related to networking components is done by MIB. The following section gives
an overview of functions performed by each components of SNMP.

3.8.1 Functions of SNMP Components

• SNMP:
6 Filtersare mathematical conditions implemented using boolean logic. These filters will
help categorizing the packets based on the output of filter.

Telecom Network Management By Nirav Patel

3.8. SNMP MANAGEMENT COMPONENTS 62

Figure 3.18: Analogy of SNMP with Computer Program

SNMP protocol is a main protocol using which a network management

application is created. SNMP defines the format of packet sent between
manager and agent and vice-versa. The packets exchanged contain the
object (variable) names and their status (values). SNMP is responsible
for reading and changing these values.

• SMI:
Network management process makes use of objects for changing the val-
ues of various network elements. These objects should have names that
are globally unique. SMI defines the rules for naming an object. It is
important to stress on the fact that SMI only defines the rules, it DOES
NOT create any object. In summary SMI functions are

– To Name the objects.

– To define the type of data that can be stored in an object.
– To show how to encode data for transmission over the network.

• MIB:
Once the rules for naming an object are clearly defined another protocol
namely the MIB is used to create an object. These objects act as the in-
terface for any network management application to manipulate properties
of network elements.

Relation between various components of SNMP Figure 3.18 gives an analogy of

network management with computer programing. In this analogy, SMI defines
the rules for naming a variable, the creating of a variable and assignment of a
value is done when the variable is initialized this is done by MIB. Finally the
logic or the interaction process between various variables of objects is defined
by SNMP protocol.

Telecom Network Management By Nirav Patel

CHAPTER 3. INTERNET MANAGEMENT (SNMP) 63

Figure 3.19: Tag Length Value (TLV) Encoding

Figure 3.20: Data Types and Tag Values

3.9 TLV Encoding Method

SMI uses another standard, Basic Encoding Rules (BER), to encode data to be
transmitted over the network. BER specifies that each piece of data be encoded
in triplet format: tag, length, and value (TLV). Once SNMO specifies the format
of message to be transmitted this data is encoded as bits that are transmitted
over the physical medium. Figure 3.19 shows the basic triplet of tag, length and
value that is used for encoding messages.
The tag is a 1-byte field that defines the type of data. Figure 3.20 shows
the data types and their tags in binary and hexadecimal numbers. The length
field is 1 or more bytes. If it is 1 byte, the most significant bit must be 0. The
other 7 bits define the length of the data. If it is more than 1 byte, the most
significant bit of the first byte must be 1. The other 7 bits of the first byte
define the number of bytes needed to define the length. The value field codes
the value of the data according to the rules defined in BER.
In the following part some examples that explain how to encode some of the
data types mentioned in figure 3.20 are explained.

3.9.1 Encoding Integer Data Type

Figure 3.21 shows how to define INTEGER 14. Note that we have used both
binary representation and hexadecimal representation for the tag. The size of
the length field is from Figure 3.20. To start with the encoding process and
INTEGER data type has a tag value 02. This becomes the first byte, which

Telecom Network Management By Nirav Patel

3.9. TLV ENCODING METHOD 64

Figure 3.21: Encoding INTEGER

Figure 3.22: Encoding OCTET STRING

indicates that the data contained in the value field is an integer. Length tag
specifies the number of bytes in the data field. Since INTEGER type data need
4 bytes, the length tag will then have the corresponding binary representation
of 4. Finally the value field contain binary equivalent of INTEGER 14.

3.9.2 Encoding Characters (OCTET STRING)

Figure 3.22 shows how to encode character string also termed as OCTET
STRING. Tag value can be obtained from Figure 3.20. ASCII values of String HI
are used for encoding. Since each character needs a 1 byte for its representation
in all 2 bytes are needed for encoding message “HI”.

3.9.3 Encoding IP Address

IP address is a main identifier of any network element. Object identifiers are
linked to a IP addresses. Encoding of an IP address 131.21.14.8 is shown in
Figure 3.23. A single IP address needs 4 bytes, this is indicated by the length
field. Each number in the IP address is encoded by its corresponding binary
representation.

Figure 3.23: Encoding IP Address

Telecom Network Management By Nirav Patel

CHAPTER 3. INTERNET MANAGEMENT (SNMP) 65

3.10 Comparison Between Different SNMP Ver-

sions
Three versions of SNMP exist: version 1 (SNMPv1), version 2 (SNMPv2), and
version 3 (SNMPv3). SNMPv1 represents the initial implementation of SNMP
that functions within the specifications of the Structure of Management Infor-
mation (SMI) and operates over protocols, such as User Datagram Protocol
(UDP) and IP. In SNMP V1 security implementation is minimum. Informa-
tion is sent over the network in form of plain text. Community name is used
to separate users who have access to SNMP messages (these users belong to a
range of restricted IP addresses). In reality these messages can be accessed by
any individual who has access to the network, as the only security mechanism
implemented is by using a community string. UDP packets can be spoofed and
any intruder can gain access to management information.
SNMP v2 is similar to SNMP v2 in implementation. The major differences
being addition of message in form of GetBulk Messages which allow the SNMP
manager to retrieve greater amount of information in a single try. Using this
message format enables manager to modify or retrieve more than one rows from
the management database. Another notable difference is the message format
of SNMP v2 is different as compared to SNMP v1. The Inform operation in
SNMPv2c enables one NMS to send trap information to another NMS and to
receive a response from the NMS.
SNMP v3 is a great improvement over the previous versions where security
is given greater importance. Features of SNMP v3

• Privacy: Encrypting data.

• Authorization: Verifying that the user allows the requested operation.

• Access control: Verifying that the user has access to the objects that
are requested.

3.11 SNMP Version 2: SNMPv2

This section describes version 2 of the protocol operations for the Simple Net-
work Management Protocol (SNMP). It provides information about the oper-
ations of the protocol with respect to sending and receiving of PDUs. It also
compares SNMPv2 with SNMP and highlights the improvements achieved in
SNMPv2. In the early 1990s work began to define SNMPv2 which would ex-
tend the capabilities of v1, and fix some of its deficiencies. However as the
protocol became less simple (initial specifications for SNMPv2 were 400+ pages
compared to the 36 pages for SNMPv1) progress slowed and some industry
members complained about the complexity. A rewritten set of RFCs was pub-
lished in late 1995 (RFC 3146). Most notably this extended the protocol to
solve some of the performance deficiencies of SNMPv1. However the working
group failed to reach agreement on the key areas of security and administrative

Telecom Network Management By Nirav Patel

3.11. SNMP VERSION 2: SNMPV2 66

framework. As a result SNMPv2 has not been accepted by the industry as a

de facto standard in the way that SNMPv1 was and work continues to define a
new standard.
The main weaknesses of SNMPv1 which SNMPv2 was supposed to address
are:
• Expressiveness of MIB definitions: As the devices and protocols to
be monitored became more complex the MIBs to describe them needed
greater expressiveness.
• Performance: When SNMP is deployed over WANs traffic overhead may
be excessive. This happens because SNMP follows a simple scalar archi-
tecture with single agent and manager. With WAN deployment there
number of hosts to be monitored on the other end of WAN may become
excessive for SNMP to handle leading to a lot of traffic(management traffic
different from data traffic).
• Security:
SNMPv1 is still primarily used only for monitoring due to the protocol’s
lack of well defined security mechanisms. The SNMPv1 specification con-
sidered the security issue, and said that agents and managers should be
grouped into Communities. An agent would identify a message as authen-
tic by verifying that it was sent by a manager in the same community
as itself. It also introduced the concepts of views and access modes onto
a MIB. However it did not specify how such schemes were to be imple-
mented. As a result the security support varied from product to product,
and inter-operability was not guaranteed.

3.11.1 SNMPv2: Improvements Over SNMPv1

The basic components of SNMPv1 and v2 are same, the functions performed by
the manager and agent are still the same.The organization model remains the
same. SNMPv2 add more number of messages to improve the communication
between management entities. It allows two managers to communicate with
each other using inform messages. Some of he improvements in SNMPv2 over
SNMPv1 are
• Bulk Messages: Get Bulk Message allows manager and agents to trans-
fer large amount of data. Initially in SNMPv1 this was accomplished by
a series of Get-request and Get Next requests.
• Manager to Manager messages: The second improvement is adding
the ability of interoperability. Version 2 allows manager to manger com-
munication using inform messages.
• MIB Enhancements: SNMP v2 added new objects to incorporate secu-
rity and SNMPv2 objects. There are also significant changes in the SNMP
groups of version one. The improvement in MIB2 for SNMPv2 is shown
in Figure 3.24

Telecom Network Management By Nirav Patel

CHAPTER 3. INTERNET MANAGEMENT (SNMP) 67

Figure 3.24: SNMPv2 Internet Group MIB structure

• Transport Mapping: SNMP has always relied on UDP for communicat-

ing messages between managers and agents. In SNMPv2 other transport
layer protocols are incorporated.

3.11.2 SNMPv2 System Architecture

System architecture of SNMPv2 is essentially same as SNMPv1, with minor
changes that allow manager to manager communication. Messages and their
functions in SNMPv2 are same as that in SNMPv1 with two new message types
added. Figure 3.25 shows the architecture of SNMPv2.
The messages GetRequest, GetNextRequest and SetRequest are the same
as in the version1 and are generated by the manager application. The message
response is also the same as that defined in version 1 and now is generated
by both the agent and the manager. The Response-PDU is generated by an
SNMP entity only upon receipt of a GetRequest-PDU, GetNextRequest-PDU,
GetBulkRequest-PDU, SetRequest-PDU, or InformRequest-PDU.
InformRequest PDU is generated by the manger application and transmit-
ted to another manager. The receiving manager responds with a Response mes-
sage. This is one of the most important improvement in SNMPv2 as this allows
communication between two managers possible. This provides interoperability
between two network management systems.
The message GetBulkRequest is generated by the manager application. It is
used to transfer large amount of data from the agent to manager. This transfer
may include transfer of an entire table of data between manager and agent. This
type of retrieval is fast and efficient.

Telecom Network Management By Nirav Patel

3.11. SNMP VERSION 2: SNMPV2 68

Figure 3.25: SNMPv2 Network Management Architecture

Telecom Network Management By Nirav Patel

CHAPTER 3. INTERNET MANAGEMENT (SNMP) 69

Figure 3.26: Bilingual Manger

SNMPv2-Trap is known as trap in SNMPv1. An SNMPv2-Trap-PDU is

generated and transmitted by an SNMP entity on behalf of a notification origi-
nator application. The SNMPv2-Trap-PDU is often used to notify a notification
receiver application at a logically remote SNMP entity that an event has oc-
curred or that a condition is present. There is no confirmation associated with
this notification delivery mechanism. The destination to which a SNMPv2-Trap
message is sent depends upon the network topology and implementation.

3.11.3 Compatibility with SNMPv1: Bilingual Manger

SNMPv2 is not backward compatible with SNMPv1. In an organization there
are sections of network that employ SNMPv1 while others may employ SNMPv2,
to provide interoperability between two such network some form of translation
mechanism is needed. This translator entity(device) converts messages from
SNMPv1 to SNMPv2 and vice versa. There are two mechanism used to im-
plement interoperability, one is proxy server(explained in section 3.5). Another
mechanism to provide interoperability is the use a bilingual manager7 .

Bilingual Manager
Bilingual manager is a device that contains modules that can interpret both
SNMPv1 as well as SNMPv2 messages. These modules are called interpreter
module. Along with interpreter module bilingual manager has a database that
contains information about the versions currently running on various agents.
This is shown in Figure 3.26
The bilingual manager does common functions of network management, this
process makes use of the version field in the SNMP message PDU to identify the
version of message. This arrangement is expensive to implement and maintain.
A suitable option for the same is provided by the proxy server.
7 Manager or device capable of understanding and translating both SNMPv1 and SNMPv2

messages

Telecom Network Management By Nirav Patel

3.12. SNMP VERSION 3: SNMPV3 70

3.12 SNMP Version 3: SNMPv3

Although SNMPv2 was released with the intention of overcoming some of the
drawbacks of SNMPv1 it never really took of as a protocol of choice. One ap-
parent reason was SNMPv2 still employed a community name based security
mechanism8 . SNMPv3 was developed to meet the need for better security in
SNMP management. SNMPv3 provides not only better security but also pro-
vides a framework for all three version of SNMP and future development in
SNMP management with minimum impact of the existing operations.
One key feature of SNMPv3 is modularization. Different functions are
grouped into different modules and these modules are implemented by the SN-
MPv3 architecture. The design of the architecture integrated the SNMPv1 and
SNMPv2 specifications with those of newly developed SNMPv3. This integra-
tion allowed the use of legacy SNMP entities by SNMPv3 agents and mangers.
Another improved feature of SNMPv3 is improved security. SNMP configura-
tion can be done remotely using secure communication links. This prevents the
messages from being modified by unwanted parties.

3.12.1 SNMPv3 Architecture

An architecture was originally defined by RFC 2271 for describing SNMP Man-
agement Frameworks. Conceptually SNMPv3 is nothing more than an extension
of SNMP to address two major areas, administration and security. A major goal
for SNMPv3, though, is to support a module architecture that can be easily ex-
tended. This way, for example, if new security protocols are advanced they
can be supported by SNMPv3 by defining them as separate modules. Before
beginning to understand SNMPv3 it important to understand the terminology
used in SNMPv3 What we used to call SNMP Agents and SNMP Managers, we
now call an SNMP entity. An SNMP entity is made up of two pieces: an SNMP
engine and SNMP applications this is shown in Figure 3.27.
As seen in the architecture SNMP engine consists of following entities
• Dispatcher:
The Dispatcher is responsible for sending and receiving messages. When a
message is received, the Dispatcher tries to determine the version number
of the message and then passes the message to the appropriate Message
Processing Model. If the version number cannot be determined the mes-
sage is discarded. Functions performed by the dispatcher can be sum-
marised as
1. Sending and Receiving messages over the network
2. Determine the correct version of SNMP message and pass it to the
correct message processing subsystem
3. Provide an interface for delivering messages from local entity to re-
mote entities
8 Only measure of security was related to community name

Telecom Network Management By Nirav Patel

CHAPTER 3. INTERNET MANAGEMENT (SNMP) 71

Figure 3.27: SNMPv3 Architecture

• Message Processing SubSystem:

The Message Processing Subsystem is made up of one or more Message
Processing Models. Figure 3.28 shows the different message processing
model incorporated inside the message processing subsystem
The SNMP message processing subsystem of SNMP engine interacts with
the dispatcher to handle SNMP messages based on their version9 . Depend-
ing upon the version of the message, the dispatcher handles the message to
the correct message processing model. A Message Processing Subsystem
that supports models for SNMPv3, SNMPv1, SNMPv2c, and Other10 .
Functions performed by the message processing subsystem can be catego-
rized as

1. Preparing messages to be sent.

2. Extracting data from received messages.

• Security Subsystem: The Security Subsystem provides security ser-

vices such as

1. Authenticating messages.
2. Encrypting/decrypting messages for privacy.
9 Version of an SNMP message is identified by the version field in the header
10 These additional models may be enterprise specific or future standards.

Telecom Network Management By Nirav Patel

3.12. SNMP VERSION 3: SNMPV3 72

Figure 3.28: Message Processing Subsystem: Message Processing Models

• Access Control Subsystem: The responsibility of the Access Control

Subsystem is straightforward: determine whether access to a managed ob-
ject should be allowed. Any application or entity that needs to access any
SNMP object will have to communicate with Access Control Subsystem.

SNMPv3 Architecture also defined different application. These application

perform different functions needed to make the SNMPv3 architecture to per-
form network management functions. These internal applications do things like
generate SNMP messages, respond to received SNMP messages, generate noti-
fications, receive notifications, and forward messages between SNMP entitites.
Currently there are five types of applications defined: (refer Figure 3.27)

1. Command Generators generate SNMP commands to collect or set

management data.
2. Command Responders provide access to management data. For exam-
ple, processing Get-Next, Get-Bulk and Set PDUs are done by a Command
Responder application.
3. Notification Originators initiate Trap or Inform messages.
4. Notification Receivers receive and process Trap or Inform messages.
5. Proxy Forwarders forward messages between SNMP entities.

Telecom Network Management By Nirav Patel

CHAPTER 3. INTERNET MANAGEMENT (SNMP) 73

Figure 3.29: SNMPv1 PDUs

Telecom Network Management By Nirav Patel

3.12. SNMP VERSION 3: SNMPV3 74

Summary
This module of Telecom Network Management summarizes the internet model
for network management. Internet model was another standard similar to OSI
network management model. The main advantage of internet based SNMP
model for network management is that it is light weight and easy to implement.
This has made SNMP a preferred choice as a network management protocol.
This module descried the various components of internet model. This descrip-
tion includes SNMP architecture and SNMP protocol specifications.
Followed by the description of the model, SNMP PDU is explained which
provides an overview of how a SNMP message is created and transmitted using
layer 3 protocols. Figure 3.29 gives all the PDUs of SNMPv1 in an organized
manner. Although SNMP is simple and easy to implement, it lacks proactive
behaviour where is can predict any faults before they happen. This is improved
by using RMON which provides data analaysis capacity. Another drawback
with SNMP protocol was the lack of security which is improved upon by later
versions namely SNMP v3. The module also covers important aspects and
improvements brought about by later modification of SNMP namely SNMPv2
and SNMPv3. The architecture of SNMP is briefly described providing an
overview of functions performed by different components. In all this modules
summaries the important aspects of internet management model and charts the
evolution of SNMP network management protocol.

Telecom Network Management By Nirav Patel

Chapter 4

Broadband Network
Management

75
4.1. INTRODUCTION 76

4.1 Introduction
Broadband network can broadly be classified in to two categories, Wide Area
Network (WAN) and access network. WAN networks are used for connecting
network that are separated by a large geographical distance. The implemen-
tation of WAN generally relies on having high speed technologies like ATM or
some form of fiber optic based technology (SONET or SDH). These technologies
are instrumental in providing high speed and high capacity data communication
which characterize broadband networks.
Telecommunication networks consists of WAN section and a local loop sec-
tion. This local loop is responsible for connecting the end user to a network.
The technology that generally form my access network is based on LAN. Which
can be implemented using cable modem, digital subscriber loop or wireless tech-
nologies.
With evolution of communication technology, boundaries between data traf-
fic, voice traffic and video traffic started fading. With advent of Integrated
Digital Subscribers Network (ISDN), the integration of these data types was
made complete. Here a single telecommunication line could carry all three
types of traffic. This led to creation of B-ISDN (Broadband ISDN). The term
broadband network has several interpretations. One such interpretation is a
network that can carry voice data and video over the same medium. One sim-
ple misconception is often that high speed data networks are called broadband
networks1 .
Improvements in the broadband networks have led to significant advances
in technologies like Asynchronous Transfer Mode (ATM), Synchronous Optical
Network (SONET), Synchronous Digital Hierarchy (SDH). ATM is one major
technology that is used for broadband networks, but this technology can be used
for any high speed network implementation. ATM is a favorable technique as it
makes use of virtual circuit switching which is essentially a packet switched tech-
nique which uses connection oriented procedures of circuit switched techniques.
Figure 4.1 shows how broadband services are used for WAN implementation
and for connecting several local loops(local networks).

4.2 Asynchronuous Transfer Mode (ATM)

Asynchronous Transfer Mode (ATM) is the cell2 relay protocol designed by
the ATM Forum and adopted by the ITU-T. ATM can be thought of as the
“highway” of the information superhighway. Some of the challenges faced by
ATM technology are

• Provide high data rate. This involves using optical fiber communication
systems for providing high speed communication.
1 The term broad refers to the mediums capability to carry signals having a large bandwidth

e.g. video signals

2 Data packets of same size

Telecom Network Management By Nirav Patel

CHAPTER 4. BROADBAND NETWORK MANAGEMENT 77

Figure 4.1: Broadband Services and Enabling Technologies

Figure 4.2: ATM TDM Multiplexing

• System must interface with existing systems with little or no degradation

in service.

• Low cost implementation of the system so that it can be accepted world

wide.

• The new system must be connection-oriented to ensure accurate and pre-

dictable delivery.

4.2.1 ATM Terminologies

• Cell:
A cell in ATM refers to a basic data unit that is used for transmission of
data.It is cell a small data unit of fixed size that is the basic unit of data
exchange in a cell network. In case of ATM information is bundled into
packets of size 53 bytes. Each of these data packet are called cells. ATM is
a cell network. All the cells have identical structure(5 bytes header and 48
bytes of information). Cells are multiplexed with other cells and routed
through a cell network. Because each cell is the same size and all are

Telecom Network Management By Nirav Patel

4.2. ASYNCHRONUOUS TRANSFER MODE (ATM) 78

Figure 4.3: Virtual Circuits in ATM

small, any problems associated with multiplexing different-sized packets

are avoided.

• Asynchronous TDM:
Data packets from individual users are separated into cells of identical size
and multiplexed using Time Division Multiplexing Scheme (TDM). Figure
4.2 shows how data from three users A, B and C are divided into smaller
cells and transmitted over a common channel using TDM scheme.

• Virtual Connection:
Information sent over the ATM network makes use of a common channel
as depicted in Figure 4.2. Connection between two points in the network is
accomplished with the help of Virtual Paths(VP), Virtual Circuit(VC) and
transmission path. Transmission path is the actual physical connection
between two points in the network. This connection is the cable or the
physical medium used for connecting end devices.
ATM makes use of virtual circuit switching technique, where each trans-
mission path is further divided into logical paths called virtual paths. Vir-
tual paths are used to carry data between nodes that belong to the same
geographical location or to nodes that are a part of the same network.
Finally each virtual path is then further divided into logical connections
called Virtual Circuits(VC). Again these circuits are not real they are log-
ical division on the virtual path that help differentiate between data from
two different user. All cells belonging to a single message follow the same
virtual circuit and remain in their original order until they reach their
destination.
Figure 4.3 shows the relation between Transmission Path, Virtual Path
and Virtual Circuits. Here the same transmission path is divided into
two virtual paths. Each virutal path is then divided into large number
of virual circuits. To facilitate virtual circuit switching, the PDU contain
two identifiers namely Virtual Path Identifier (VPI) and Virtual Circuit
Identifier (VCI). Each connection between two end users is differentiated
by using two VPI and VCI. In Figure 4.3 TP is divided into two VPI
numbered 14 and 18 and VPI 14 is further divided into VCI numbered

Telecom Network Management By Nirav Patel

CHAPTER 4. BROADBAND NETWORK MANAGEMENT 79

21, 32 and 45. A single communication stream between two end users is
identified by a combination of (VPI,VCI).
• ATM Layers:
ATM is a WAN communication technology that enables long distance
communication. This makes it imperative to incorporate different tech-
nologies and different protocols at the application layer. Data from these
applications is then converted to a fixed data size of 53 bytes and is then
transmitted over the physical network. To allow this flexibility ATM in-
corporates ATM Application Adaptation Layers (AAL). There are 4 types
of adaptation layers shown in Figure 4.4. ATM defined 3 layers where the
top most layer is AAL. The application adaptation layer (AAL) is used
only by the end points.
Next layer used by ATM is the ATM layer, this layer provides routing,
traffic management, switching, and multiplexing services. It processes
outgoing traffic by accepting 48-byte segments from the AAL sublayer.
The addition of a 5-byte header transforms the segment into a 53-byte
cell. A cell is 53 bytes in length with 5 bytes allocated to header and 48
bytes carrying payload (user data may be less than 48 bytes). Most of the
header is occupied by the VPI and VCI. Figure 4.5 shows the functions
performed by ATM layer where the data received from the upper layers is
segmented into cells of equal size.
• ATM Packet:
In the ATM hierarchy, AAL receives data from different applications and
protocols and converts it into a form that can be used by the ATM layer.
ATM layer converts the data received from AAL into segments of 48 bytes
each and then attaches a header of 5 bytes. This header is shown in Figure
4.6. Header contains the information about VPI and VCI. A combination
of VPI and VCI is used to identify a unique data stream over a transmis-
sion path.
– GFC(Generic Flow Control): This bit is used to provide flow
control and congestion control to all the data segments that are trans-
mitted.
– Virtual Path Identifier(VPI): The 8 bits of VPI are used to divide
a physical transmission path into logical paths. Each path is then
identified by a number.
– Virtual Circuit Identifier(VCI): 16 bit field of the VCI helps
distinguish different data-stream on the same virtual path. Each
VPI refers to a connection between two end devices.
– Payload Type(PT): Payload type is used to distinguish of provide
information about the type of data that is being carried by the cell.
– Header Error Control: These bits are added by the ATM layer to
provide error protection to the transmitted information.

Telecom Network Management By Nirav Patel

4.2. ASYNCHRONUOUS TRANSFER MODE (ATM) 80

Figure 4.4: ATM layers : Application Adaptation Layer

Figure 4.5: ATM layer Functions

Figure 4.6: ATM Header Components

Telecom Network Management By Nirav Patel

CHAPTER 4. BROADBAND NETWORK MANAGEMENT 81

4.3 ATM LAN Emulation

LAN Emulation is a software protocol allowing ATM networks to behave like
local area networks, running existing applications unchanged. It also allows
connectivity between ATM networks and existing LANs. LANs can have high
data rates upto 1Gbps, but this predominantly makes use of IP for transmission.
This makes it difficult to interface them with technologies like SONET and SDH,
which predominantly makes use of fiber optics. ATM-LAN emulation provides a
solution to this problem. LANE (LAN ATM Emulation) allows ATM networks
to run LAN application without any modifications. So in effect the user gets to
use LAN applications without being worried about the interfacing of LAN with
underlying ATM network. Some advantages that LANE offers users are
• High-speed LANs is a major ATM application. The extra bandwidth (155
instead of 10 Mbps) and the dedicated links appeal to many users.
• To achieve widespread deployment, ATM must allow todays LAN applica-
tions to run over the new ATM hardware, making migration from todays
Ethernet and Token Ring LANs to tomorrows ATM LANs easier.
The introduction of a new technology such as ATM may require significant
costs, unless it is performed in a stepwise fashion. There are two aspects to an
economical migration: hardware costs (new network interface cards, equipment,
cabling, test equipment); and software costs (change or rewrite of existing ap-
plications, additional management tools). LAN Emulation is designed to help
ease the migration. It is a software protocol running over ATM equipment that
offers two major points:

• The ability to run all existing LAN applications over ATM without change.
The immediate benefit is that users do not have to reinvest in applications.
• The ability to interconnect ATM equipment and networks to existing
LANs, and to link logically separate LANs via one ATM backbone. The
advantage is that ATM equipment only has to be introduced where it is
needed.

LAN Emulation is being defined by the ATM Forum in the LAN Emulation
workgroup. Note LAN Emulations greatest advantage is also its greatest dis-
advantage: it hides the ATM layer from the application. This means that
applications running over LAN Emulation cannot use the additional benefits of
an ATM network: support for multiple kinds of data, a specifiable Quality of
Service, priority and congestion management and more.

4.3.1 LAN Emulation Architecture

LAN Emulation is a protocol stack or a software that runs over ATM network
and allows users to enjoy LAN services over ATM network. Figure 4.7 shows
the protocol architecture that enables LAN Emulation.

Telecom Network Management By Nirav Patel

4.3. ATM LAN EMULATION 82

Figure 4.7: ATM LAN Architecture

LAN Architecture shows an overlay of LANE over ATM protocol stack.

There are some basic differences between ATM and LAN communication.

• ATM is connection oriented communication.

• ATM makes one-to-one connection between two workstation, while a LAN

relies on multicast and broadcast process for establishing a communication
link.

• Communication needs ATM address (20 Bytes) while LAN relies on phys-
ical address on the NIC.

As shown in Figure 4.7 LANE (LAN Emulation) layer is above AAL which
allows interfacing of LAN applications with the underlying ATM network. AAL
is followed by ATM and physical layer. As shown in the Figure 4.7 lower three
layers namely physical, ATM and AAL represent the base ATM architecture on
which the LAN architecture3 is implemented with the help of LAN Emulation
Layer.
The ATM adaptation layer interfaces LAN with ATM and allows LAN to
run varied types of applications. LUNI (LAN Emulation User Network Inter-
face) is specified by ATM Forum. LUNI provides interface between the user
and the LAN emulation layer. Before understanding LANE, it is necessary to
understand some terms used to describe various network elements. They are
described in the next section.
3 Network Layer, LLC are a part of LAN.

Telecom Network Management By Nirav Patel

CHAPTER 4. BROADBAND NETWORK MANAGEMENT 83

4.3.2 Components of ATM LAN Emulation Architecture

There are several participants in the LAN emulation (LE) protocol opera-
tion: the LAN Emulation Client (LEC), the LAN Emulation Server (LES), the
LAN Emulation Configuration Server (LECS), and the Broadcast and Unknown
Server (BUS). Each of these is described below.

• LAN Emulation Client (LEC):

The LEC is the user requiring LAN emulation services. Typically, it is the
workstation running the application or the ATM bridge which connects the
ATM network with the legacy LAN. There can be many LAN Emulation
Clients in an emulated LAN. LE is any client that uses the service, this
client is generally a server that is accessed by many LAN users.

• LAN Emulation Server (LES):

The LES implements address registration (allowing stations to register
their MAC and ATM addresses) and provides address resolution (answers
ARP (Address Resolution Protocol) requests by converting between MAC
and ATM addresses). Each emulated LAN can have only one LES. How-
ever, a physical LAN can serve several emulated LANs, each with its own
LES.

• LAN Emulation Configuration Server (LECS):

The LECS provides configuration information, including the address of
the LE server, the type of emulated LAN and the maximum frame size.
Each network can only have one LECS.

• Broadcast/Unknown Server (BUS):

The BUS performs all broadcasts and multicasts. Broadcast and multi-
casts are performed in case the information has to be transmitted to all the
stations. The other situation where the BUS will broadcast the message
is when a source LEC has sent an ARP4 to the LES, and does not wish
to wait for a response before starting the data transfer to the destination
LEC.

While the ATM Forum specifies that the are three separate logical compo-
nents to the LAN Emulation service (the LES, LECS and BUS), it deliberately
does not specify whether they are physically separate or united. This decision
is left to the vendors. Many vendors merge the LES, LECS and BUS into a
single physical unit. This implementation is shown in Figure 4.8 where all the
components are implemented on a single device.
Four functions are performed by the LUNI interface shown in Figure 4.8
namely (a) Initialization (b) registration (c) address resolution and (d) data
transfer.
4 Address Resolution Protocol: Protocol used over LAN for finding the physical address

corresponding to an IP address.

Telecom Network Management By Nirav Patel

4.3. ATM LAN EMULATION 84

Figure 4.8: LUNI Interface and LANE Components

• Initialization:
Initialization involves obtaining ATM addresses of LAN Emulation Server
that are available on the network. This is the server that is running the
application that enables LANE. Second function involves joining/leaving
a particular ATM network.

• Registration:
Registration informs the LES of the individual MAC addresses that the
LEC will communicate with or those LEs who will use the services of LEC.

• Address Registration:
As explained in the previous sections ATM makes use of 20 Byte ATM
Address where as LAN relies on physical address for routing and data
delivery. Address registration enables correct resolution of LAN address
onto ATM addresses.

• Data Transfer:
Data transfer involves transmission of data between source and destina-
tion. This is done by converting data into frames and then encapsulating
them as ATM frames. AAL-55 is the protocol used to perform LAN Em-
ulation. Transmitted data is then decapsulated at the receiving end.

4.3.3 Communication In ATM LAN Emulation

In networking terms having a pure ATM based communication system is of-
ten very costly and seldom implemented. A more tradition implementation is
shown in Figure 4.9, ATM equipment is interconnected with legacy network
technologies. This implementation is cost effective and more practical.
In this network, only selected stations requiring the higher bandwidth asso-
ciated with ATM are connected with ATM links. A typical example of such a
station is the file server, where a great deal of network traffic is concentrated.
5 There are different types of ATM Adaptation layer, AAL-5 has protocols that enable

implementing LANE.

Telecom Network Management By Nirav Patel

CHAPTER 4. BROADBAND NETWORK MANAGEMENT 85

Figure 4.9: ATM Connected to Legacy LAN

The other stations remain connected via Ethernet and thus retain their original
network cards.
Figure 4.10 shows a typical ATM LAN Emulation network which makes use
of different components of LANE and communicates with legacy systems like
Ethernet.
In the Figure 4.10 when the ATM workstation wants to communicate with
LAN it functions as LEC (LAN Emulation Client) and interfaces with the LAN
through LUNI(LAN User to Network Interface). On the LAN side interfacing
of LAN part of the network with ATM is done through bridge. The bridge that
has both technologies: a LAN port towards the legacy LAN and an ATM port
towards the newer ATM network.
The LEC requires various services to function correctly in LANE environ-
ment. These services are provided by components namely LES, BUS and LECS.
Though they are three different components their implementation can be done
on a single network device(switch). There are two types of connections estab-
lished in ATM communication namely Data Connection and Control Connection
(shown in Figure 4.10).
Any communication happening in ATM happens via virtual circuit, which
is a connection oriented service. The same is not applicable for LAN or con-
ventional Ethernet LAN. To facilitate this communication one LEC(ATM side)
communicates to a device situated on LAN with the help of LAN Emulation
Server(LES). In ATM devices are identified using 20 byte ATM address while
on the LAN side of the network devices are identified using a 6 Byte MAC
address. This address translation is enabled by LES. Clients have to register
with LES, so while communicating with other devices the correct ATM/MAC
address can be resolved and a connection oriented communication can be estab-
lished. This communication can happen between various hosts, in-case the host
on LAN it may send an Ethernet or Token Ring packet. To facilitate commu-

Telecom Network Management By Nirav Patel

4.3. ATM LAN EMULATION 86

Figure 4.10: ATM LAN Emulation Connection using LUNI

nication with an ATM device, Ethernet packet is encapsulated as an ATM cell

and transmitted. This transmission is done using virtual circuit.
Conventional LANs also allow broadcast and multicast of packets. This is
not possible in ATM as all the communication is always connection oriented6 .
To enable multicast and broadcast7 features in LANE, a device called BUS
(Broadcast/Unknown Server) is used. A broadcast packet is send from a LEC
to the BUS, which then converts this packet into a series of unicast messages,
which are sent to various LECs. The address of the LECs is attached in the
packet, this enables the BUS to send series of one to one communication to all
the LECs. The unknown server function of the BUS is used when the LEC
cannot obtain the address translation from the LES. Figure 4.10 shows the
communication happening between ATM host and BUS.
Different segment of Emulated LAN are controlled by LES, so assigning a
host to a different LAN can be done by making changes at the LES, this avoids
any need for physical changes. This is one advantage of LANE, which also helps
in implementation of Virtual LANs.

6 Source and destination are first connected via a virtual circuit, that remains dedicated to

them for the period of communication.

7 Multicast and Broadcast are a type of one to many connection, where one hosts commu-

nicates or sends packets to more than one destination.

Telecom Network Management By Nirav Patel

CHAPTER 4. BROADBAND NETWORK MANAGEMENT 87

4.4 Virtual LAN (VLAN)

In traditional LAN architecture all the devices connected to a single switch come
under the same LAN. This kind of an architecture creates certain problems.
First problem being that of broadcast messages. Any broadcast for a particular
LAN segment will de delivered to all the devices. This increases the data traffic
in a LAN segment as well as devices keep responding to unwanted messages.
Second and a more important concern with a traditional LAN architecture is
that with security. Since any device connected to the switch automatically
becomes part of the LAN, data over the LAN is accessible to it. This raises
certain security concerns where critical data can simply be accessed by physically
connecting a device to a LAN Port (port on a switch).
To overcome this drawback the concept of Virtual LAN is implemented.
VLAN is a logical grouping of users and resources connected to ad-
minstratively defined ports on a switch. What this definition simply im-
plies is VLAN allows different users that are connected to the same switch to
be placed in different LANs. These LANs are defined by the administrator and
each port on the switch can be placed in a different LAN. Such administrative
segregation or logical division of a physical LAN into a logical LANs is the
fundamental concept of VLAN. A VLAN is treated a separate LAN segment
and broadcasts generated for a section of LAN is limited only to those users
belonging to that VLAN.
Figure 4.11 gives the pictorial representation of a VLAN where multiple
users connected to the same switch are a part of different LAN. These LANs
are administratively defined and named as VLAN2- VLAN7. Each VLAN can
be assigned to a different section of a company. Here separation of users is done
based on the VLAN to which they belong. In this way the users need not be
in the same physical location but they can still be a part of the same VLAN.
Inter VLAN communication has to be accomplished using a router. So in effect
creating a VLAN is same as creating different networks.
In ATM emulation the feature of VLAN can be implemented. This imple-
mentation can be done by administratively defining VLANs and these VLANs
are implemented on LAN Emulation Configuration Server(LECS). Figure 4.12
shows implementation of VLAN on LECS. Each port og the switch is assigned to
a particular VLAN. This creates virtual separation of different users. Commu-
nication between two users happens just like any normal ATM communication.

Some of the advantages that VLAN offers are

• VLANs enhance security greatly. Some of the users that wish to have high
level of security can be logically grouped into a single VLAN.

• VLAN groups users logically based on the functions or services that they
want. This grouping makes them independent of physical or geographical
location.

• Broadcast messages are restricted to limited number of users.

Telecom Network Management By Nirav Patel

4.4. VIRTUAL LAN (VLAN) 88

Figure 4.11: VLAN Implementation

Figure 4.12: ATM VLAN Configuration

Telecom Network Management By Nirav Patel

CHAPTER 4. BROADBAND NETWORK MANAGEMENT 89

Figure 4.13: ATM Reference Architecture Overview

Figure 4.14: ATM Interface Reference Architecture

4.5 ATM Reference Model

ATM reference model encompasses the connection of private networks with pub-
lic ATM networks. Figure 4.12 shows an overview of ATM reference model where
private ATM networks are connected to public ATM networks via UNI (User
Network Interface). Figure 4.14 depicts the ATM network management refer-
ence model specified for end-to-end ATM network management as specified by
ATM Forum. This figure shows the connection between various network man-
agement systems and the ATM networks they manage. For end-to-end ATM
network management communication between private and public NMSs is im-
portant. This is made possible with the help of M1, M2, M3 and M4 interfaces.

The five key interfaces that enable communication between different NMS

Telecom Network Management By Nirav Patel

4.6. INTERIM LOCAL MANAGEMENT INTERFACE(ILMI) 90

for efficient end-to-end ATM network management are

• M1: The management interface between the private management system

and the end ATM device (i.e. between ATM workstation and the man-
agement system).

• M2: The management interface between the private management system

and the switches that make up the local private ATM network.

• M3: The management interface the private management system and the
public management system.

• M4: The management interface between the switches in the public man-
agement system and the public ATM network.

• M5: The management interface between two public management systems

from two different carriers.

Each of these interfaces are used for management of different aspects of ATM
network (covered in later sections).

4.6 Interim Local Management Interface(ILMI)

Until proper standards are formalized by ITU and ATM-Forum for management
of local ATM networks, SNMP becomes the preferred choice for implementing
local management. The Simple Network Management Protocol (SNMP) and an
ATM UNI Management Information Base (MIB) will be required to provide
any ATM user device with status and configuration information concerning
the Virtual Path and Channel Connections available at its UNI(User Network
Interface). The overall involvement of ILMI in ATM network management can
be illustrated by Figure 4.15, where the connection between NMS and the end
user using ATM network services are depicted.
This interface depicts various links present between various components on
the ATM network. ILMI helps the network management system to retrieve
configuration and status information of various interfaces and devices8 .
A simplified view of the ILMI interface is shown in Figure 4.16.
Some of the features of ILMI protocol are

• The ILMI communication protocol is an open management protocol (i.e.,

SNMP/AAL initially).

• The ILMI communication takes place between adjacent ATM UMEs.

• Interim Local Management Interface (ILMI) functions for a UNI provide

status, configuration, and control information about link and physical
layer parameters at the UNI.
8 UME =UNI Management Entity. An interface connecting user to the ATM network.

Telecom Network Management By Nirav Patel

CHAPTER 4. BROADBAND NETWORK MANAGEMENT 91

Figure 4.15: ILMI Interface for ATM Management

Figure 4.16: Simplified View of ILMI Interface

Telecom Network Management By Nirav Patel

4.7. ATM INTERFACES 92

• The Simple Network Management Protocol (SNMP) without UDP and IP

addressing along with an ATM UNI Management Information Base (MIB)
were chosen for the ILMI.

• An Interim Local Management Interface (ILMI) supports bi-directional

exchange of management information between UNI Management Entities
(UMEs) related to UNI ATM layer and physical layer parameters.

• The communication across the ILMI is protocol symmetric.

Figure 4.17 shows a general network topology implementing UNI at different

interfaces.
ILMI protocol was developed as a set of interim specifications for ATM
network management it makes use of both SNMP MIB for network management
(MIB-2) and MIB structure developed by ATM-Forum. Figure 4.18 shows the
MIB structures used for ATM management.
The role and importance of SNMP in ATM management is made more rele-
vant by use of interface M1 and M2 which make use of SNMP MIB while com-
municating with the private NMS. Interface M2 has a proxy agent that allows
translation of ATM MIB (ILMI MIB) to SNMP MIB thus allowing management
of ATM network with the help of SNMP.

4.7 ATM Interfaces

The following subsections give a brief introduction on the functions performed
by M1-M5 network management interfaces.

4.7.1 M1 Interface: Management of ATM Network Ele-

ment
M1 interface is the management interface between the private management sys-
tem and the end ATM device (i.e. between ATM workstation and the manage-
ment system). NMS of local networks(refer Figure 4.14) make use of SNMP for
management, M1 interface allows management of ATM end devices(Network
Elements) by the local NMS. This is depicted in Figure 4.19 that shows a net-
work element(end device) which is an ATM device. This device has a SNMP
agent installed on it, allowing it to communicate with the local NMS.
When an ATM end device communicates with the ATM network there are
only three possible modes of communication that can be implemented in the
physical layer. They are DS1(1.544 Mbps T1 Line) DS2(44.736 Mbps using T3
Line) and SONET (nx 155.52 Mbps). Only one of the three possible physical
layer implementation can be done. So for simplicity, depending upon the type of
physical layer implementation only one of the transmission groups9 are installed
in the agent. These transmission groups are shown in Figure 4.20.
9 Objects of MIB that enable managing the physical layer transmission.

Telecom Network Management By Nirav Patel

CHAPTER 4. BROADBAND NETWORK MANAGEMENT 93

Figure 4.17: Examples of Equipment Implementing ATM UNI and ILMI

Telecom Network Management By Nirav Patel

4.7. ATM INTERFACES 94

Figure 4.18: ATM Management MIB Structure

Figure 4.19: M1 Interface: SNMP Management

Figure 4.20: Transmission Groups for ATM

Telecom Network Management By Nirav Patel

CHAPTER 4. BROADBAND NETWORK MANAGEMENT 95

Figure 4.21: ATM M2 Interface

4.7.2 M2 Interface: Management of Private Network

M2 is the management interface between the private management system and
the switches that make up the local private ATM network. Figure 4.21 shows
the M2 management interface.

4.7.3 M3 Interface: Customer Network Management

M3 is also called as the customer network management interface. M3 describes
the interface between the customer and the carrier management systems. This
gives a customer a view of carrier’s network. M3 allows customers to supervise
the use of their portion of public ATM network.

4.7.4 M4 Interface: Public Network Management

M4 is an interface needed to manage public network services. This involves both
management of Network Elements(NE) and management of services provided
over the network. This interface enables the public NMS to view properties
of network elements and its properties. The main complication arises with
interoperability. Public networks make use of OSI based TMN (CMIP Common
Management Information Protocol) while local networks make use of SNMP.
The interface M4 is specified in such a way that both the SNMP and CMIP
based approach can be used for management.

Telecom Network Management By Nirav Patel

4.7. ATM INTERFACES 96

4.7.5 M5 Interface: Inter NMS Communication

M5 is the most complicated interface as it involves communication between
NMSs of different carriers. There is no standard available for this interface yet.

Summary
This section describes the basics of Broadband Network Management tech-
niques. Broadband network consists of various interacting technologies like
LAN at the local network and ATM/SONET/SDH for implementing WAN.
Since ATM is one prime technology used for implementing WAN, this module
revolves around management of ATM networks. In the first few sections, basics
of ATM communication are introduced. Where ATM is a connection oriented
communication technique. Main advantage of ATM is the high speed communi-
cation provided with the help of underlying physical layer implementations like
SONET and SDH. To leverage the advantages of ATM for Local LANs the con-
cept of ATM LAN Emulation is introduced. Finally the ATM reference model
and its interfaces are described to sum up to broadband management process

Telecom Network Management By Nirav Patel

Chapter 5

Network Management
Applications

In the previous few chapters the concept or the idea of network management
revolved around managing the communication network. This involved manage-
ment of network elements and the interconnections between various networks.
This module adds another dimension to the concept of network management.
The model of network management is extended to include Business manage-
ment and Service management. Figure 5.1 presents this integration of busi-
ness management and service management. In this figure the highest level of
management is in form of business management. These functions are the activi-
ties related to running an enterprise (educational institutes, Telecommunication
service providers, IT organizations). These activities rely of a second level of
management which is the service management. Service management involves
managing various aspects of services provided by the enterprise. These services
refer to the communication related service provided by the enterprise. Figure
5.2 shows the relation between service management and various lower layer on
Telecom Management Network.
These management activities will in turn rely on a third level of service called
network management and system management. Network management and Sys-
tem management aggregate data obtained from the lowest layer of of actual
network elements (devices) and system resources. This module deals with net-
work management applications as defined by Functional Model of OSI Network
management model. These application are the functions performed by a net-
work management systems i.e. Fault Management, Configuration Management,
Accounting Management, Performance Management and Security Management
(FCAPS). A brief introduction of FCAPS is presented in section 2.7. (refer that
section before further study)
Network management and service management are not only related to tech-
nical aspects of a telecom network, but they also have a bearing on the formula-
tion of business decisions and policy making procedures. Based on the statistics

97
5.1. CONFIGURATION MANAGEMENT 98

Figure 5.1: Network Management: Business and Service Management

and analysis performed in the process of Network management and service man-
agement policies are created. This policies are create measures and metrics for
measuring the performance of a network and in extension the services provided
by the service provider.

5.1 Configuration Management

Configuration Management is responsible for setting various parameter in net-
work components. In case of a failure component management performs the
task of informing the team responsible for provisioning that service. Such fail-
ure are recorded and maintained in a database which can later be reviewed for
analysis. Configuration Management is used in context of discovering a network
topology and mapping network. At the NOC the systems engineer needs to have
a view of the physical network that is being managed. Along with the topology,
configuration of the various network elements should also be available for quick
and efficient troubleshooting of any fault. This is the scope of configuration
management. Some of the functions that are performed as part of configuration
management are
• Network Provisioning
• Network Topology View
• Inventory Management

Telecom Network Management By Nirav Patel

CHAPTER 5. NETWORK MANAGEMENT APPLICATIONS 99

Figure 5.2: TMN Conceptual Architecture: Service Management

Telecom Network Management By Nirav Patel

5.1. CONFIGURATION MANAGEMENT 100

5.1.1 Network Provisioning

Provisioning in network management implies making a service available as and
when demanded by the user. This involves providing communication links and
paths for data transmission. In computer communication networks provisioning
requirements are different as compared to circuit switched communication. In
computer networks data is usually communicated using packet switched tech-
niques. In connection less packet switching when a source initiates communi-
cation with a destination packets are route over the network. Here each packet
takes different route based on the network congestion and routing policies. Pro-
visioning of networks implies that links and switching paths should be available
when the demand for them arises. In case of congestion data packets should be
stored in buffers and retransmitted when network congestion is reduced.
This also includes using virtual circuit switching techniques based on the
demand. In case of broadband management or WLAN management, ATM
technology is used which relies on the use of Virtual Circuit Switching. Network
Provisioning in case of broadband management will involve providing VPIs and
VCIs to various connections initiated for data communication.

5.1.2 Inventory Management

Inventory management provides the engineers at NOC with the configuration
information of all devices on the network. Engineers at NOC rely heavily on this
for timely trouble shooting and fault management. Configuration management
refers to having a repository that contains the configuration information of all
the components in the network. The database contains location and network
addresses of all hardware devices as a well as information about programs,
versions and updates installed in the network components.
There are three types of configurations, ‘static configuration’ (permanent
configuration), this is the configuration from which the network would start if
its started from an idle status. Second configuration is ‘current configura-
tion’ this is the configuration on which the network is currently running. It is
not necessary that current running configuration is same as static configuration.
Third type of configuration is the ‘planned configuration’, this refers to the
configuration to which the network will be upgraded in future. The NOC has a
display that reflects the dynamic configuration of network and its status. Any
changes in the configuration are reflects in the NMS tool used by administra-
tors. In case of a service failure some temporary changes may be added to a
components, these dynamic change are reflected in NMS tool.
Initially hierarchical or scalar database were used for maintaining configura-
tion information. This made it difficult to add any new entry. These databases
were than replaced by relational databases which are easier to modify. With the
use of OSI based network management model, object oriented philosophy was
used in management process. An object-oriented relational database is ideally
suited for configuration and inventory management.

Telecom Network Management By Nirav Patel

CHAPTER 5. NETWORK MANAGEMENT APPLICATIONS 101

5.1.3 Network Topology

Effcient network management relies on having up to date information about
the topology of network. To make the process of network management more
efficient, knowledge of network topology is also important, this is done by main-
taining a physical topology of the managed network. One major difficult is
having updated information of the topology. In case the network shrinks or
expands, network topology should be automatically updated. This difficult is
overcome by network management system by using auto-discovery mechanism.
There are various ways of mapping network topology one such technique is
auto-discovery using the broadcast ping. Here all the devices that respond to
the broadcast ping send by the network management system are mapped. Along
with this broadcast ping, more information can be obtained about the devices
by using SNMP GET REQUESTS. Although this process is simple, broadcast
ping creates great amount of management traffic, which may not be a good idea
at times. To avoid large number of broadcasts, the NMS can also look up the
arp table of network switches to collect information about the network topology.
Second technique to maintain the physical topology is the arp messages sent
out to network elements. The devices that are on will respond to these messages
allowing the NMS to be map them onto the physical topology.
Practical difficulties in generating dynamic topology is that sometimes due
to implementation of VLAN, the logical topology will differ from the physical
topology. This is solved by having VLAN grouping based on IP Address or
MAC Address.

5.2 Fault Management

Service failure or unavailability of any IT service is a common fault with net-
work management. Restoring a service for users is a responsibility of NOC. This
involves detection and isolation of problem that caused service failure. In some
situations, network will do this automatically. Such networks are called self-
healing networks. But in some situations when automatic recovery or healing is
not possible, Network Management Systems (NMS) are used to detect failures.
These failures can be in form of component failure and a NMS raises appropri-
ates alarms and sends notifications to the concerned team. Fault management
is a five step process.

• Fault Detection

• Fault Location

• Service Restoration

• Root cause analysis

• Problem Resolution

Telecom Network Management By Nirav Patel

5.3. SECURITY MANAGEMENT 102

When a service fails or any device malfunctions an alarm is raised to the

network management system. The most important part of fault management
is service restoration. Root cause analysis of the fault or problem resoultion
takes a lower priority. When a service failure occurs a ticket is generated ei-
ther automatically (NMS has that option) or manually by an engineer. This
ticket contains detailed description about the problem. All the possible avail-
able information about service failure event is present in the ticket. This ticket
is then assigned to I&M group for actual maintenance or repair work if nec-
essary. Generally any ticket contains tracking number of the request, affected
service, affected user, any SLA that is associated with the disrupted service.
Once service has been restored and the underlying problem resolved, ticket is
closed.

5.2.1 Fault Detection

Fault detection process is generally automated in any complex network manage-
ment process. There are two techniques to implement fault detection. Polling
method and Trap generation. An application program sends regular ping com-
mands and awaits for response. If a predefined number of ping request fail,
connectivity to that device is declared broken and NMS is informed about the
failure. To optimize network performance, the number of ping messages and the
frequency of ping messages is balanced with data traffic.
Trap generation is another method where a device upon failure informs the
NMS, SNMP V1 makes use of TRAP message to implement fault detection.

5.2.2 Fault Location and Isolation

Fault location is an important step in resolving a fault and restoring a service.
One thing to notice is that a single fault in the network can affect multiple
services at the same time raise multiple different alarms. It is important to
identify the source of failure to restore services. Fault location is generally
implemented by using the network topology as a guide map. The origin of
the problem could be traced by walking down the topology tree to where the
problem starts. Once the source of failure has been located diagnostic tool and
procedures are used to resolve the problem.

5.3 Security Management

Any communication networks main concern is having proper and efficient secu-
rity management. Any network management process has to ensure the following
three security concerns
• Secure Communication
• Secure Access to data
• Physical security of network devices

Telecom Network Management By Nirav Patel

CHAPTER 5. NETWORK MANAGEMENT APPLICATIONS 103

These three security concerns shape various policies and procedures instituted
under the process of security management. This section talk about such policies
and procedures that are implemented to ensure secure communication for all the
users. Along with policies and procedures this section also gives an introduction
on the equipments needed to implement these security measures.

5.3.1 Policies and Procedures

IETF defines security policies as “ A formal statement of rules by which people
who are given access to an organization’s technology and information assets
must abide”. When it comes to access of information, all the employees should
not have access to all the resources of network. This is enabled by creating
accounts for employees that can access only a subset of the available network
resources. Such security policies and procedures try to answer the following
questions
1. Identify what resource is to be protected (applications, hardware)
2. What do the above mentioned resources have a threat from ?
3. How likely are the threats?
4. Implement measures to counter those threats.
5. Review the process continuously and make improvements if weakness are
found
Threats to any network are
1. Unauthorized access to resources and/or information
2. Unintented and/or unauthorized Disclosure of information
3. Denial of service
Any security policy is created to counter these general threats. Once threat
assessment and policy making process is completed, the next step involves de-
ploying suitable equipments that will enable these policies. Some of these com-
ponents are described in the next sections

5.3.2 Firewall
Firewalls are just another tool in the quest for system security. They provide a
certain level of protection and are, in general, a way of implementing security
policy at the network level. The level of security that a firewall provides can
vary as much as the level of security on a particular machine. There are the
traditional trade-offs between security, ease of use, cost, complexity, etc.
Main task of a firewall is to protect a network from external threats. It
monitors and controls traffic flowing through the network. Firewall functions
can be implemented on a router or a gateway. Same functionalities can be

Telecom Network Management By Nirav Patel

5.3. SECURITY MANAGEMENT 104

achieved using a stand alone device. In corporate network firewalls are located
at the border of the network, where the private network meets the internet.
Along with hardware components, firewall functionalities are also implemented
on hosts in form of software application that protects the hosts from malwares.

Telecom Network Management By Nirav Patel

Chapter 6

Telecommunication
Management Network

The term TMN is introduced by the ITU-T (the former CCITT) as an abbrevi-
ation for Telecommunications Management Network. The concept of a TMN is
defined by Recommendation M.3010. TMN has a strong relationship with OSI
management, and defines a number of concepts that have relevance for Internet
Management.
According to recommendation M.3010 Telecommunication Network Manage-
ment (TMN) is a conceptually different network from the communication net-
work. This management network interfaces with the communication network
at multiple places and helps manage the underlying communication network
efficiently. Figure 1.1 shows the relation between TMN and communication
network (the same fig presented here for ease of reference).
According to Figure 6.1, the interface points between the TMN and the
telecommunication network are formed by Exchanges and Transmission systems.
For the purpose of management, these Exchanges and Transmission systems
are connected via a Data Communication Network to one or more Operations
Systems. The Operations Systems perform most of the management functions;
these functions may be carried out by human operators but also automatically.
A TMN provides management functions for telecommunication networks and
services and offers communications between itself and the telecommunication
networks, services and other TMNs. The basic concept behind a TMN is to
provide an organized architecture to achieve the interconnection between various
types of Operations Systems (OSs) and/or telecommunications equipment for
the exchange of management information using an agreed architecture with
standardized interfaces including protocols and messages.
The objective for the TMN specifications is to provide a framework for
telecommunications management. By introducing the concept of generic net-
work models for management, it is possible to perform general management
of diverse equipment, network and services using generic information models

105
6.1. DIFFERENCE BETWEEN TMN AND COMMUNICATION
NETWORKS 106

Figure 6.1: General Relationship Between TMN and Telecommunication Net-

work

and standard interfaces. A TMN is intended to support a wide variety of

management areas which cover the planning, installation, operations, admin-
istration, maintenance and provisioning of telecommunications networks and
services. Recommendation

6.1 Difference Between TMN and Communica-

tion Networks
A TMN can vary in complexity from a very simple connection between an OS
and a single piece of telecommunications equipment to a complex network in-
terconnecting many different types of OSs and telecommunications equipment.
A TMN may provide management functions and offer communications both
between the Operations Systems(OS) themselves, and between OSs and the
various parts of the telecommunications network. A TMN may also offer com-
munication between two TMNs of different networks in order to support the
management of international and national telecommunications networks.
A telecommunications network consists of many types of analogue and dig-
ital telecommunications equipment and associated support equipment, such
as transmission systems, switching systems, multiplexes, signalling terminals,
front-end processors, mainframes, cluster controllers, file servers, etc. When
managed, such equipment is generically referred to as network elements (NEs).
Figure 6.1 shows the general relationship between a TMN and a telecommu-
nications network which it manages. A TMN is conceptually a separate net-

Telecom Network Management By Nirav Patel

CHAPTER 6. TELECOMMUNICATION MANAGEMENT NETWORK 107

Figure 6.2: TMN Architecture Components

work that interfaces a telecommunications network at several different points to

send/receive information to/from it and to control its operations. A TMN may
use parts of the telecommunications network to provide its communications.

6.2 TMN Architecture

A TMN architecture consists of three components as shown in Figure 6.2

• Functional Architecture

• Physical Architecture

• Informational Architecture

6.2.1 Functional Architecture

The TMN functional architecture is a structural and generic framework of man-
agement functionality that is subject to standardization. Figure 6.3 shows build-
ing blocks of functional architecture of TMN.
In this figure, two types (OSF and MF) are completely drawn within the
box labelled ‘TMN’. This way of drawing indicates that these function blocks
are completely specified by the TMN recommendations. The other three types
(WSF, NEF and QAF) are drawn at the edge of the box to indicate that only
parts of these function blocks are specified by TMN.
The TMN functional architecture introduces the concept of reference point to
delineate function blocks. Five different classes of reference points are identified.
Three of them (q, f and x) are completely described by the TMN recommenda-
tions.
Function blocks connected by conceptual interfaces, called reference point
as shown in Figure 6.4

• x: Interface between operations systems that belong to different domains;

e.g., interface between two NMSs belonging to two different domains

• q3: Interface between two OSFs in the same domain

Telecom Network Management By Nirav Patel

6.2. TMN ARCHITECTURE 108

Figure 6.3: Blocks of Functional Architecture of TMN

Figure 6.4: Reference Point in TMN architecture

• qx: Interface between mediation function such as RMON and agent in the
network element
• f: Interface to the workstation

Figure 6.5 shows interconnections between different functional blocks along

with the interfaces used to communicate the information.

• Operations System Functions: The Operations System Functions

(OSF) block initiates management operations and receives notifications.
In terms of the manager-agent model, the OSF may be seen as the man-
ager specific functions. An OSF communicates with the NEF over a q3
reference point. Functions performed by Operations systems: e.g., NMS,
testing, accounting, trouble tracking.
• Network Element Functions: A typical telecommunication network
consists of exchanges and transmission systems. In TMN terminology,
exchanges and transmission systems are examples of network elements
(NEs). The functions that are performed by NEs are Network Element
Functions (NEFs). According to TMN, these functions include:
1. Primary (or telecommunications) functions. These functions are the
subject of management and support the exchange of data between the
users of the telecommunication network.

Telecom Network Management By Nirav Patel

CHAPTER 6. TELECOMMUNICATION MANAGEMENT NETWORK 109

Figure 6.5: Interface Between Different Functional Blocks

Figure 6.6: Q Adaptor Functions

2. Management functions, which allow the NEF block to operate in an

agent specific role.

• Work Station Functions: The Work Station Function (WSF) block

provides the means to interpret TMN information for the management
information user. The WSF includes support for interfacing to a human
user (at the g reference point). Such aspects of support are not considered
to be part of the TMN.

• Q Adaptor Functions: The Q Adaptor Function (QAF) block is used to

connect to the TMN those entities which do not support standard TMN
reference points. Figure 6.6 shows how QAF interconnects non TMN
entities to TMN entities.

• Mediation Functions: The Mediation Function (MF) block is located

within the TMN and acts on information passing between NEFs or QAFs,
and OSFs. A MF block can be used to connect a single, as well as multiple
NEFs and QAFs to an OSF.

Telecom Network Management By Nirav Patel

6.2. TMN ARCHITECTURE 110

Figure 6.7: Physical Architecture of TMN

6.2.2 Physical Architecture

Next to a functional architecture, TMN also defines a physical architecture.
The latter architecture shows how TMNs functions, which were defined by the
functional architecture, can be implemented into physical equipment. TMNs
physical architecture is thus defined at a lower abstraction level than TMNs
functional architecture, Figure 6.7 shows relation of functional architecture with
physical architecture.
A physical architecture of TMN elaborates how the functions specified by
the functional architecture have to be implemented using physical blocks. As an
example, function architecture specifies the need for communication between dif-
ferent block, the physical architecture will then specify the block or components
needed to establish this communication. To enable this functionality physical
architecture makes use of some building blocks. These building blocks will have
a name similar to some functional blocks.

Building Blocks of Physical Architecture

TMNs Physical Architecture defines the following building blocks:
• Network Element
• Mediation Device (MD).
• Q Adaptor (QA).
• Operations System (OS).
• Work Station (WS).
• Data Communication Network (DCN).
Building blocks always implement the function blocks of the same name
(e.g. Network Elements perform Network Element Functions, Mediation Devices
perform Mediation Functions etc.). It is possible to implement multiple function
blocks (of the same or of a different type) into a single building block. The

Telecom Network Management By Nirav Patel

CHAPTER 6. TELECOMMUNICATION MANAGEMENT NETWORK 111

Figure 6.8: Relation Between Different Components of TMN Architecture

Operations System, for example, may be used to implement multiple OSFs, but
may also be used to implement an OSF, MF and a WSF. In the case a building
block implements multiple function blocks of different types, ”the choice on the
building blocks name is determined by the predominate usage of the block”.
A special kind of building block is the Data Communication Network (DCN).
As opposed to the others, this building block does not implement any TMN func-
tion block. In fact, the DCN is used by other building blocks for the exchange
of management information; the DCNs task is to act as a transport network.

6.2.3 Informational Architecture

Management of telecommunication network as a whole relies on communica-
tion of proper information between managing entities. It can be thought of as
an information processing application, where the information communicated by
various devices have to be understood correctly by the applications. To effec-
tively manage complex networks and support network operator/service provider
business processes, it is necessary to exchange management information between
management applications implemented in multiple managing and managed sys-
tems. Thus telecommunication management is a distributed application.
The TMN information architecture, in order to promote interoperability,
is based on standardized open management paradigms that support the stan-
dardized modelling of the information to be communicated. TMNs information
architecture uses an object oriented approach and is based on OSIs Management
Information Model. According to this model, entities to be managed are viewed
as objects. These objects are described in terms of

• Attributes, which are the properties or characteristics of the object.

• Operations, which are performed upon the object.

• Behaviour, which is exhibited in response to operations.

• Notifications, which are emitted by the object.

Telecom Network Management By Nirav Patel

6.3. TMN LOGICAL LEVEL ARCHITECTURE 112

Figure 6.9: Informational Architecture

Figure 6.10: Hierarchy of TMN Architecture

Figure 6.9 shows how any managed entity is described in using TMN informa-
tional architecture.

6.3 TMN Logical Level Architecture

TMN makes use of hierarchical architecture for implementing interoperability.
Such hierarchies can be described in terms of management layers; the architec-
ture that describes this layering is called the Logical Layered Architecture. To
deal with the complexity of management, the management functionality with
its associated information can be decomposed into a number of logical layers.
The principle of such layering is shown in Figure 6.10.
The advantage of having layered architecture is that each layer manager
passes on relevant information to the manager above it. Same applies for man-
agers at all level. This layered architecture makes use of principle of abstrac-
tion1 . Manager at each layer has an abstract view of layer above it and layer
below it. This makes designing different layers easy. Layers are created by
grouping similar set of functions into a single layer. A usual decomposition of
management functionality leads to the following layers of abstraction:
1 Only display the relevant information, while the actual working is hidden from the user

Telecom Network Management By Nirav Patel

CHAPTER 6. TELECOMMUNICATION MANAGEMENT NETWORK 113

Figure 6.11: Logical Layered Architecture

• Element management layer.

• Network management layer.

• Service management layer.

• Business management layer.

This logical layered architecture is shown in Figure 6.11

6.3.1 Element Management Layer

The functions of individual Network Elements are managed by Operations Sys-
tems Functions (OSF) in the Element Management layer. This layer deals with
vendor specific management functions and hides these functions from the layer
above, the Network Management layer. This layer is related to functions per-
formed by individual components in the physical architecture.
Examples of functions performed at the Element Management layer are:

• detection of equipment errors,

• measuring power consumption,

Telecom Network Management By Nirav Patel

6.3. TMN LOGICAL LEVEL ARCHITECTURE 114

• measuring the temperature of equipment,

• measuring the resources that are being used, like CPU-time, buffer space,
queue length etc.,

• logging of statistical data,

• updating firmware.

6.3.2 Network Management Layer

Whereas the responsibility of the Element Management layer is to manage NEFs
implemented within single pieces of equipment, the responsibility of the Network
Management layer is to manage the functions related to the interaction between
multiple pieces of equipment.
Examples of functions performed at this layer are:

• creation of the complete network view,

• creation of dedicated paths through the network to support the QoS de-
mands of end users,

• modification of routing tables,

• monitoring of link utilization,

• optimizing network performance, and

• detection of faults.

6.3.3 Service Management Layer

The Service Management layer is concerned with management of those aspects
that may directly be observed by the users of the telecommunication network.
These users may be end users (customers) but also other service providers (ad-
ministrations). Service Management builds upon the management information
that is provided by the Network Management layer, but does not ‘see’ the in-
ternal structure of the network (principle of abstraction).
Examples of functions performed at the Service Management layer are:

• Quality of Service management (delay, loss, etc.),

• Accounting,

• Addition and removal of users,

• Address assignment,

• Maintenance of group addresses.

Telecom Network Management By Nirav Patel

CHAPTER 6. TELECOMMUNICATION MANAGEMENT NETWORK 115

6.3.4 Business Management layer

The Business Management layer is responsible for the management of the whole
enterprise. This layer has a broad scope; communications management is just
a part of it. Business management can be seen as goal setting, rather than
goal achieving. For this reason business Management can better be related to
strategical and tactical management, instead of operational management, like
the other management layers of TMN.

6.4 TMN Implementation

TMN architecture consists of three components namely Functional Architec-

ture, Informational Architecture and Physical Architecture. The functional ar-
chitecture framework permits the specification of what functions have to be
achieved in the TMN implementation. The information architecture permits
the specification of what information (i.e. data) has to be stored so that the
functions defined in the functional architecture can be achieved in the TMN
implementation. Physical architecture translates the requirements of functional
and informational architectures into actual network implementation. This is the
implementation of TMN architecture. The TMN implementation, that meets
the requirements of the TMN functional and information specifications, may
vary greatly from one TMN solution to another. TMN implementations are not
currently a subject for standardization.
TMN implementations have to balance a number of diverse constraints such
as cost, performance, and legacy deployments, as well as incorporate new func-
tionality. Since every TMN implementation will have different sets of these
constraints to cope with, reality dictates that there will be many physical ar-
chitecture implementations. These implementation architectures are the result
of different distributions of the fundamental elements.
Figure 6.12 shows a protocol independent implementation of TMN takes
shape. There are many possible and different distributions. Each implementa-
tion has to satisfy the needs identified and expressed in both the TMN functional
architecture and the TMN information architecture specifications.
This is a very general description of TMN implementation. In reality each
implementation of TMN will have involve very specific components based on
the business requirements. Each component of TMN is mapped into specific
blocks. This is shown in Figure 6.13. Requirements of business are categorized
into different functions which are grouped into functional blocks. Based on the
TMN specifications and requirement of functional and informational blocks, the
physical architecture is implemented with proper selection of network elements.

Telecom Network Management By Nirav Patel

6.4. TMN IMPLEMENTATION 116

Figure 6.12: TMN Implementation (Protocol Independent)

Figure 6.13: TMN Architecture Implementation

Telecom Network Management By Nirav Patel

CHAPTER 6. TELECOMMUNICATION MANAGEMENT NETWORK 117

Summary
This module introduces the general idea of Telecommunication Management
Networks. Here TMN is a standard that aims to go beyond simple network
management. TMN aims to achieve business management and to do this various
standardization process are adopted. The module introduces the concept of
TMN architecture and describes various functional blocks involved in TMN
architecture. Finally provides a introduction of the layered architecture and
various ways to implement TMN for efficient business management.

Telecom Network Management By Nirav Patel

6.4. TMN IMPLEMENTATION 118

Telecom Network Management By Nirav Patel

List of Figures

1.1 General Relationship Between TMN and Telecommunication Net-

work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.2 Network Management Functional Groups and Activities . . . . . 10
1.3 Network Management Flowchart . . . . . . . . . . . . . . . . . . 11
1.4 Domains of Network Management Architecture . . . . . . . . . . 15
1.5 Network Management Dumbbell Architecture . . . . . . . . . . . 15
1.6 Network Management Components . . . . . . . . . . . . . . . . . 16
1.7 Network Management Interoperability . . . . . . . . . . . . . . . 16

2.1 Network Management Functions Overview . . . . . . . . . . . . . 21

2.2 OSI Network Management Model . . . . . . . . . . . . . . . . . . 25
2.3 Two Tier Network Management Model . . . . . . . . . . . . . . . 26
2.4 Three Tier Network Management Model . . . . . . . . . . . . . . 27
2.5 Network Management Model with MoM . . . . . . . . . . . . . . 28
2.6 Network Management Model with MIB . . . . . . . . . . . . . . 29
2.7 Comunication Model in Network Management . . . . . . . . . . . 30
2.8 Protocol used in Communication Model . . . . . . . . . . . . . . 31
2.9 Functional Model Network Mangement . . . . . . . . . . . . . . . 32

3.1 Protocol Hierarchy in SNMP based NMS . . . . . . . . . . . . . 39

3.2 Hierarchy of Devices in SNMP . . . . . . . . . . . . . . . . . . . 41
3.3 SNMP Mangement Architecture and Protocol Stack . . . . . . . 42
3.4 Two-Tier Organization Model . . . . . . . . . . . . . . . . . . . . 44
3.5 Two Tier Organization Model with Two Managers . . . . . . . . 44
3.6 Three Tier Organization Model with RMON . . . . . . . . . . . 45
3.7 Organization Model with Proxy Server . . . . . . . . . . . . . . . 45
3.8 SNMP Community with Multiple Manager/Agent Pair . . . . . . 48
3.9 SNMP Community . . . . . . . . . . . . . . . . . . . . . . . . . . 49
3.10 SNMP Access Policy Single Community . . . . . . . . . . . . . . 50
3.11 SNMP Administration Model . . . . . . . . . . . . . . . . . . . . 51
3.12 Encapsulation of SNMP PDU . . . . . . . . . . . . . . . . . . . . 52
3.13 SNMP Get/Set Type PDU . . . . . . . . . . . . . . . . . . . . . 53
3.14 SNMP Proxy Server and Firewall . . . . . . . . . . . . . . . . . . 55
3.15 SNMP Proxy Server SNMPv1-v2 . . . . . . . . . . . . . . . . . . 56

119
LIST OF FIGURES 120

3.16 RMON Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

3.17 Components of SNMP . . . . . . . . . . . . . . . . . . . . . . . . 61
3.18 Analogy of SNMP with Computer Program . . . . . . . . . . . . 62
3.19 Tag Length Value (TLV) Encoding . . . . . . . . . . . . . . . . . 63
3.20 Data Types and Tag Values . . . . . . . . . . . . . . . . . . . . . 63
3.21 Encoding INTEGER . . . . . . . . . . . . . . . . . . . . . . . . . 64
3.22 Encoding OCTET STRING . . . . . . . . . . . . . . . . . . . . . 64
3.23 Encoding IP Address . . . . . . . . . . . . . . . . . . . . . . . . . 64
3.24 SNMPv2 Internet Group MIB structure . . . . . . . . . . . . . . 67
3.25 SNMPv2 Network Management Architecture . . . . . . . . . . . 68
3.26 Bilingual Manger . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
3.27 SNMPv3 Architecture . . . . . . . . . . . . . . . . . . . . . . . . 71
3.28 Message Processing Subsystem: Message Processing Models . . . 72
3.29 SNMPv1 PDUs . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

4.1 Broadband Services and Enabling Technologies . . . . . . . . . . 77

4.2 ATM TDM Multiplexing . . . . . . . . . . . . . . . . . . . . . . . 77
4.3 Virtual Circuits in ATM . . . . . . . . . . . . . . . . . . . . . . . 78
4.4 ATM layers : Application Adaptation Layer . . . . . . . . . . . . 80
4.5 ATM layer Functions . . . . . . . . . . . . . . . . . . . . . . . . . 80
4.6 ATM Header Components . . . . . . . . . . . . . . . . . . . . . . 80
4.7 ATM LAN Architecture . . . . . . . . . . . . . . . . . . . . . . . 82
4.8 LUNI Interface and LANE Components . . . . . . . . . . . . . . 84
4.9 ATM Connected to Legacy LAN . . . . . . . . . . . . . . . . . . 85
4.10 ATM LAN Emulation Connection using LUNI . . . . . . . . . . 86
4.11 VLAN Implementation . . . . . . . . . . . . . . . . . . . . . . . 88
4.12 ATM VLAN Configuration . . . . . . . . . . . . . . . . . . . . . 88
4.13 ATM Reference Architecture Overview . . . . . . . . . . . . . . . 89
4.14 ATM Interface Reference Architecture . . . . . . . . . . . . . . . 89
4.15 ILMI Interface for ATM Management . . . . . . . . . . . . . . . 91
4.16 Simplified View of ILMI Interface . . . . . . . . . . . . . . . . . . 91
4.17 Examples of Equipment Implementing ATM UNI and ILMI . . . 93
4.18 ATM Management MIB Structure . . . . . . . . . . . . . . . . . 94
4.19 M1 Interface: SNMP Management . . . . . . . . . . . . . . . . . 94
4.20 Transmission Groups for ATM . . . . . . . . . . . . . . . . . . . 94
4.21 ATM M2 Interface . . . . . . . . . . . . . . . . . . . . . . . . . . 95

5.1 Network Management: Business and Service Management . . . . 98

5.2 TMN Conceptual Architecture: Service Management . . . . . . . 99

6.1 General Relationship Between TMN and Telecommunication Net-

work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
6.2 TMN Architecture Components . . . . . . . . . . . . . . . . . . . 107
6.3 Blocks of Functional Architecture of TMN . . . . . . . . . . . . . 108
6.4 Reference Point in TMN architecture . . . . . . . . . . . . . . . . 108
6.5 Interface Between Different Functional Blocks . . . . . . . . . . . 109

Telecom Network Management By Nirav Patel

LIST OF FIGURES 121

6.6 Q Adaptor Functions . . . . . . . . . . . . . . . . . . . . . . . . . 109

6.7 Physical Architecture of TMN . . . . . . . . . . . . . . . . . . . . 110
6.8 Relation Between Different Components of TMN Architecture . . 111
6.9 Informational Architecture . . . . . . . . . . . . . . . . . . . . . . 112
6.10 Hierarchy of TMN Architecture . . . . . . . . . . . . . . . . . . . 112
6.11 Logical Layered Architecture . . . . . . . . . . . . . . . . . . . . 113
6.12 TMN Implementation (Protocol Independent) . . . . . . . . . . . 116
6.13 TMN Architecture Implementation . . . . . . . . . . . . . . . . 116

Telecom Network Management By Nirav Patel