St.

Mary’s University School of Graduate Studies

Department Of Computer Science

Quality of Service Comparison of Seamless Multi -Protocol Level

Switching and Multi-Protocol Level Switching Networks

By
Genet Daba
Advisor
Dr. Asrat Mulatu

A Thesis Submitted to St. Mary’s University, Department Of Computer Science In

Partial Fulfillment Of The Requirement For The Award Of Master Degree Of
Science In Computer Science

March, 2022
Addis Ababa, Ethiopia
 DECLARATION

I, the undersigned, hereby declare that this thesis entitled: “Quality of Service Comparison of
Seamless Multi-Protocol Level Switching and Multi-Protocol Level Switching Networks”
is my original work and that all sources of materials used for this thesis have been duly
acknowledged. This work has been submitted partially, or in full, by any other person for an
award of a degree in any other University or Institution and carried out the study under the
guidance and support of the research advisor of Dr. Asrat Mulatu. The Assistance and help
received during the course of this investigation have been duly acknowledged.

----------------------------------- ----------------------------------- -----------------------------------

Researcher’s Name Signature Date

The thesis has been submitted for examination with my approval as an Advisor

----------------------------------- ----------------------------------- -----------------------------------

Advisor’s Name Signature Date

i
 THESIS APPROVAL FORM

This is to certify that the thesis prepared by (Genet Daba), entitled “Quality of Service
Comparison of Seamless Multi-Protocol Level Switching and Multi-Protocol Level
Switching Networks” and submitted in partial fulfillment of the requirements for the
Degree of Masters of Science in Computer Science MSc complies with the regulations of
the University and meets the accepted standards with respect to originality and quality.

APPROVED BY BOARD OF EXAMINERS:

--------------------------------- -------------------------------- ----------------------------------

Candidate Signature Date

--------------------------------- -------------------------------- ----------------------------------

Advisor Signature Date

Signature of Board of Examiners

--------------------------------- -------------------------------- ----------------------------------

External Examiners Signature Date

--------------------------------- -------------------------------- ----------------------------------

Internal Examiners Signature Date

--------------------------------- -------------------------------- ----------------------------------

Chairman / School Dean Signature Date

ii
 Abstract
A seamless MPLS network is one in which Multi-Protocol Level Switching is used for all packet
forwarding within the network, from the time a packet enters the network until it leaves it.
Seamless Multi-Protocol Level Switching was created with the goal of providing an architecture
that can handle a wide range of services on a single Multi-Protocol Level Switching. Access,
aggregation, and the core network are all integrated into one platform.

The goal of traffic engineering is to make network operations more effective and dependable
while also maximizing network resource consumption and traffic performance. Because of the
high cost of network infrastructure and the commercial and competitive nature of the Internet,
traffic engineering has become an essential function in many big Autonomous Systems. These
issues highlight the need of maximizing operational efficiency.

Traffic oriented performance objectives include the aspects the Quality of Service of traffic
streams. In a single class, best effort Internet service model, the key traffic-oriented performance
objectives include: minimization of packet loss, minimization of delay, minimization of jitter and
maximization of throughput.

The primary goals of Quality of Service are bandwidth management, controlled jitter, latency
and improved packet loss characteristics to provide satisfactory services for users.

The goal of this thesis is to improve the quality of service on multi-protocol level switching that
is seamless. Two scenarios are used to examine the influence on Quality of Service parameters:
one with Seamless Multi-Protocol Level Switching and the other with Resource Reservation
Protocol -Traffic Engineering Seamless Multi-Protocol Level Switching. To compare the
performances of the two situations, simulation tools such as Graphical Network Simulator-3,
Ostinato, Paessler Router Traffic Grapher, and excel are utilized. On various Quality of Service
metrics, the result demonstrates that Resource reservation protocol Seamless Multi-Protocol
Level Switching is superior than Seamless Multi-Protocol Level Switching.

Keywords: MPLS, Seamless MPLS, QoS, Traffic Engineering, RSVP, Network Analysis

iii
 TABLE OF CONTENTS
DECLARATION ................................................................................................................................................ i
THESIS APPROVAL FORM .............................................................................................................................. ii
Abstract ........................................................................................................................................................ iii
LIST OF ABBREVIATION ............................................................................................................................... vii
LIST OF FIGERS .............................................................................................................................................. x
LIST OF TABLES ............................................................................................................................................. xi
Introduction .................................................................................................................................................. 1
1.1 Background of the Study ........................................................................................................................ 1
1.2 Statement of the Problem ...................................................................................................................... 2
1.3 Objectives ............................................................................................................................................... 3
1.3.1 General Objective ..................................................................................................................... 3
1.3.2 Specific Objectives ................................................................................................................... 4
1.4 Methodology .......................................................................................................................................... 4
1.5 Scope of the Study.................................................................................................................................. 4
1.7. Limitation of the Study ................................................................................................................. 5
1.8. Thesis Layout .......................................................................................................................................... 5
2.1 Introduction to MPLS ............................................................................................................................. 6
2.1.1 Forwarding Equivalence Class (FEC) ....................................................................................... 6
2.1.2 Label Switching Router (LSR) ................................................................................................. 7
2.1.3 Label Switched Path (LSP) ....................................................................................................... 7
2.1.4 Label Distribution Protocols (LDP) .......................................................................................... 8
2.2 Label Distributions .................................................................................................................................. 8
2.2.1 Piggyback the Labels on an Existing IP Routing Protocol ....................................................... 9
2.2.2 Separate Protocol for Label Distribution .................................................................................. 9
2.3 Control Plane and Forwarding Plane ...................................................................................................... 9
2.3.1 Control Plane .......................................................................................................................... 10
2.3.2 Data Plane ............................................................................................................................... 10
2.6. Seamless MPLS ..................................................................................................................................... 10
2.6.1 Deterministic End-to-end Service Restoration........................................................................ 10
2.6.2 Decoupled Network and Service Architectures ...................................................................... 11
2.6.3 Service Flexibility with Simplified Provisioning and Operations .......................................... 11
2.6.4 Building Scalable Networks.................................................................................................... 11

iv
2.7. Seamless MPLS Architecture ................................................................................................................ 11
2.8. Traffic Engineering................................................................................................................................ 13
2.8.1 Refresh Reduction Techniques ............................................................................................... 13
2.8.2 RSVP with Traffic Engineering Extensions ........................................................................... 13
2.8.3 RSVP PATH Message ............................................................................................................ 14
2.8.4 RSVP RESERVATION Message ........................................................................................... 14
2.8.5 RSVP error Message ............................................................................................................... 14
2.8.6 RSVP Tear Message ............................................................................................................... 14
2.9. Quality of Service (QoS)........................................................................................................................ 15
2.9.1. Through put ............................................................................................................................. 15
2.9.2. Delay ....................................................................................................................................... 15
2.9.3. Jitter......................................................................................................................................... 16
2.9.4. Packet Loss ............................................................................................................................. 16
2.10. Review of Related Works ................................................................................................................ 17
2.10.1 Review of Related Works Summary ....................................................................................... 20
CHAPTER THREE .......................................................................................................................................... 22
SIMULATION RESULTS AND ANALYSIS ........................................................................................................ 22
3.1 Overview of Simulation Tools............................................................................................................... 22
3.1.1 Graphical Network Simulator-3 (GNS3) ................................................................................ 22
3.1.2 VMware Workstation.............................................................................................................. 22
3.1.3 GNS3 VM ............................................................................................................................... 22
3.1.4 Cisco Internetworking Operating System (IOS) images ......................................................... 23
3.1.5 Ostinato ................................................................................................................................... 23
3.1.6 Paessler Router Traffic Grapher (PRTG) ................................................................................ 23
3.2 Simulation Scenarios and Network Topology ...................................................................................... 23
3.3 Seamless MPLS Network topology details ........................................................................................... 24
3.4 Router interface configuration ............................................................................................................. 24
3.5 Router configuration for IP network .................................................................................................... 26
3.6 Router configuration for Seamless MPLS network .............................................................................. 26
3.7 Simulation Parameters Analysis ........................................................................................................... 27
3.7.1 Throughput Analysis ............................................................................................................... 27
3.7.2 Latency Analysis ..................................................................................................................... 29
3.7.3 Packet Loss Analysis .............................................................................................................. 30

v
 3.7.4 Jitter Analysis.......................................................................................................................... 31
3.8. Comparative Analysis of Results .......................................................................................................... 33
CHAPTER FOUR ........................................................................................................................................... 35
CONCULUSIONS AND FUTRURE WORKS ..................................................................................................... 35
4.1 CONCLUSIONS ...................................................................................................................................... 35
4.2 CONTRIBUTIONS ................................................................................................................................... 35
4.3 FUTURE WORKS .................................................................................................................................... 36
References .................................................................................................................................................. 37
Appendix: Scripts for Seamless MPLS Configuration and RSVP-TE Seamless MPLS Configuration ............ 40

vi
 LIST OF ABBREVIATION
ARP Address Resolution Protocol

ATM Asynchronous transfer mode

BFD Bidirectional Forwarding Detection
BGP Border Gateway Protocol
BGP Border Gateway Protocol
BSC Base station controller
CPU Central processing unit
DOD Downstream-on-Demand
FEC Forwarding Equivalence Class
DSL Digital Subscriber Line
IGP Interior Gateway Protocol
PC Personal computer.
IP Internet protocol
DSLAM Digital Subscriber Line Access Multiplexer
ENSP Enterprise Network Simulation Platform
FTTB Fiber to the building
FTTC Fiber to the Curb
FTTH Fiber to the home
GNS3 Graphical Network Simulator-3
HTTP Hypertext Transfer Protocol.
IBGP Internal Border Gateway Protocol
ICMP Internet Control Message Protocol
IETF Internet Engineering Task Force
IGMP Internet Group Management Protocol
IPV4 Internet Protocol Version 4
IPV6 Internet Protocol Version 6
ISIS Intermediate System to Intermediate System

vii
ISP Internet service provider
LDP Label distribution protocol
LER Label Edge Router
LFA Loop-free alternate
LFIB Label Forwarding Information Base
LIB Label Information Base
LSP Label switching path
LSR Label Switching Router
MPLS Multiprotocol Label Switching
MTU Maximum transmission unit
NLRI Network Layer Reachability Information
NNTP Network News Transfer Protocol
NSR Nonstop active routing
OAM Operations, administration and management
OLT Optical line terminal
OSI Open Systems Interconnection
OSPF Open Shortest Path First
PRTG Paessler Router Traffic Grapher
QoS Quality of service
RFC Request for Comments
RIP Routing Information Protocol
RNC Radio network controller
RR Route Reflector
RSVP Resource reservation protocol
RTSP Real Time Streaming Protocol
SIP Session Initiation Protocol
SLA Service level agreement
TCP Transmission Control Protocol

viii
TCP/IP Transmission Control Protocol/Internet Protocol,
TDP Tag Distribution Protocol
TE Traffic engineering
UDP User Datagram Protocol
VM Virtual Machine
VPN Virtual private network
WAN Wide area network

ix
 LIST OF FIGERS

Figure 2.1: Position of MPLS in OSI model .................................................................................................... 6

Figure 2.2: MPLS Packet Header (6) .............................................................................................................. 7
Figure 2.4: MPLS with multiple domains [4] ............................................................................................... 12
Figure 3.1: Graph of throughput for scenarios 1 and 2 .............................................................................. 28
Figure 3.2: PRTG Graph of throughput for scenarios 1 and 2 .................................................................... 28
Figure 3.3: Graph of latency for scenarios 1 and 2 .................................................................................... 30
Figure 3.4: Graph of packet loss for scenarios 1 and 2 .............................................................................. 31
Figure 3.5: Graph of Jitter for scenarios 1 and 2........................................................................................ 32

x
 LIST OF TABLES

Table 2.1: Quality standards for throughput .............................................................................................. 15

Table 2.1: Quality standards ITU-TG114 for delay ...................................................................................... 16
Table 2.3: Quality standards ITU-T G 114 for Jitter .................................................................................... 16
Table 2.4: Quality standard for packet lose ................................................................................................ 16
Table 1.1: Summary of Review of Related Works ....................................................................................... 20
Table 3.1: Topology details ......................................................................................................................... 24
Table 3.2: Router interface details.............................................................................................................. 25
Table 3.3: IP network details....................................................................................................................... 26
Table 3.4: MPLS network details ................................................................................................................. 27
Table 3.5: Seamless MPLS-TE tunnel details ............................................................................................... 27
Table 3.6: Unified MPLS components ......................................................................................................... 27
Table 3.7: Output of Throughput for Seamless MPLS and RSVP TE-Seamless MPLS at different file sizes 28
Table 3.8: Output of latency for Seamless MPLS and RSVP TE-Seamless MPLS at different file sizes ....... 29
Table 3.9: Output of packet loss for Seamless MPLS and RSVP-TE Seamless MPLS at different file sizes . 30
Table 3.10: Output of jitter for Seamless MPLS and RSVP-TE Seamless MPLS at different file sizes ......... 32

xi
 CHAPTER ONE
Introduction
1.1 Background of the Study
Multiprotocol Label Switching (MPLS) is displacing other WAN technologies due to its
improved reliability and efficiency. It is preferable than traditional Internet Protocol (IP) routing,
which bounces data all across the internet before sending it to its final destination. They are more
dependable in terms of developing scalability, secure platforms, efficient platforms, and service
level agreement verification [3]. Asynchronous transfer mode (ATM) is a switching technique
used by telecommunication networks that uses asynchronous time-division multiplexing to
encode data into small, fixed-sized cells. For WAN connectivity, various technologies such as
Frame Relay, ATM, T1 or E1 dedicated links were used in the past. Layer2 VPNs were used in
business networks that were not scalable to maintain security issues. The MPLS VPN offers
scalability and can be used to separate larger organizations. Multiprotocol Label Switching
(MPLS) is data forwarding technology that increases the speed and controls the flow of network
traffic. With MPLS, data is directed through a path via labels instead of requiring complex
lookups in a routing table at every stop. [2].

Label Switching protocols are increasingly used in packet-based computer networks for traffic
engineering and other purposes. Label Switching routers (LSRs) create label switched paths in a
label switching network using the MPLS signaling protocols. (LSPs). MPLS protocols are used
by LSRs to accept MPLS label map pings from downstream LSRs and to advertise MPLS label
map pings.

MPLS provides an established and reliable network foundation for core and aggregation
networks. MPLS can also be used in access networks, such as telephone or Digital Subscriber
Line (DSL) backhaul networks. MPLS has two layers: the Transport Layer and the Service Layer
(for MPLS VPNs, for example). In both cases, the protocols and encapsulation are the same [47].

Although the encapsulation is identical, the application of MPLS varies, especially in terms of
signaling, control plane, provisioning, scalability, and update frequency. At the service layer,
only service-specific information is shared, and each service can theoretically use its own design
and protocols. The services are operated using the transport layer. 5A seamless MPLS network

Page 1
uses MPLS for all packet forwarding within the network, from the time a packet enters to the
time it leaves [40].

Seamless MPLS was created with the aim of supporting a broad range of services on a single
MPLS platform that completely integrates connectivity, aggregation, and the core network. The
design of the building Seamless MPLS enables network and service providers to install service
creation points practically anywhere in the network, allowing for more flexible service and
service creation. Without the need for dedicated service creation areas on fixed sites, service
creation can be achieved based on existing requirements. The development of services is easier
with Seamless MPLS' versatility [33].

The ability of a network to deliver better service to specified network traffic through multiple
technologies is referred to as Quality of Service (QoS). The main purpose of QoS is to provide
dedicated bandwidth, controlled jitter and latency (which is required by some real-time and
interactive traffic), and improved loss characteristics. It's also crucial to ensure that giving one or
more flows precedence does not cause other flows to fail. QoS technologies provide the
fundamental building blocks for future business applications in campus, wide-area network, and
service provider networks. Whether it's a small company network, an Internet service provider,
or an enterprise network, almost every network can benefit from QoS for maximum
efficiency [6].

Quality of service (QoS) is the use of mechanisms or technologies that work on a network to
control traffic and ensure the performance of critical applications with limited network capacity.
It enables organizations to adjust their overall network traffic by prioritizing specific high-
performance applications [22]. When utilizing MPLS, however, you have an additional option
for implementing QoS for labeled packets. A signaled channel through the network between two
routers is known as an LSP. The label on top of the packet can be used to indicate a portion of
the QoS for that packet.

1.2 Statement of the Problem

The use of real-time application like (multimedia) in Seamless MPLS are becoming increasingly
significant However, most of the routing strategies used in seamless MPLS provide only best
effort service.

Page 2
MPLS routing protocol also provides best effort service without any guarantee of QoS
requirements. The main problems or limitation of Seamless MPLS is no link optimization, it’s
difficult to arrange bandwidth utilization and maintain the traffic path during congestion. Due to
this limitation delivering better QoS for user is difficult. Real-time applications need QoS, since
they are, by nature, highly time sensitive to reach destination within minimal delay compared to
other traffics.

To address these challenges and improve QoS, traffic engineering can be applied to the core,
aggregation and access layers in seamless MPLS. In this thesis work, it is planned to improve the
QoS of Seamless MPLS networks by applying resource oriented traffic engineering techniques.

This is due to the fact that BGP does not allow for the deployment of QoS across various
domains (inter-domain or inter-AS). Another drawback is that, by its very nature, BGP has a
delayed convergence time (about 30 seconds) when the network experiences a breakdown.
Because the lack of QoS in the inter domain network has an impact on the overall QoS,
implementing QoS in the intra-domain network alone does not guarantee end-to-end QoS
throughout the entire network.

To deal with these issues, Seamless MPLS is one convergent inter-domain network design
proposed to improve management, service provisioning, and scalability, but its impact on QoS
parameters has not been validated or quantified to ensure end-to-end QoS assurances. Despite the
fact that some academics have undertaken traffic analysis for a single MPLS domain, the author's
understanding of the influence of Seamless MPLS on QoS is limited. This analysis is necessary
to determine the benefits and drawbacks of replacing existing multi-domain MPLS with
Seamless MPLS based on QoS criteria. Limitation of MPLS they require less powerful routers
with limited capabilities, require the customer to control routing, they support only IP traffic and
PE routers are underutilized

1.3 Objectives

1.3.1 General Objective

The general objective of this thesis is to quality of service in seamless Multiprotocol Level
Switching Networks by applying resource reservation protocol.

Page 3
 1.3.2 Specific Objectives
 Identify the Seamless MPLS limitations with influence on TE.
 To compare and Analysis the current seamless MPLS architecture with the traffic
engineering applied seamless MPLS architecture by Throughput, Latency, Packet
Loss and Jitter.
 Identify the current state of communication techniques through TE.
 Identify the features in seamless MPLS which promise improved QoS.
 To simulate and evaluate Seamless MPLS architecture.
 To assess the impact of Seamless MPLS on QoS parameters.

1.4 Methodology
Seamless MPLS architecture & implementation scenarios with its benefits compared to Traffic
Engineering applied seamless MPLS architecture are explained. In the implementation part, a
practical environment is developed using network simulation tool, Enterprise Network
Simulation Platform (ENSP) or GNS3, and two scenarios are built in order to collect test results
from the simulator.

In this thesis state-of-the-art, related works and statement of the problem are used as baseline to
achieve the objectives. The methodology starts with investigating different technologies enabling
Seamless MPLS architecture. Then the methods of simulating and evaluating the architecture
with QoS perspective are followed. A theoretical study of seamless MPLS and QoS features are
done thoroughly along with the evaluation of the limitations of the seamless MPLS architecture.

The two scenarios are built in such a way that first an ordinary network is built with a Seamless
MPLS network. Then the same network topology is implemented with traffic engineering
Seamless MPLS features and the test results are collected from the simulator using Network
Quality Analyzer (NQA) technology for the two scenarios. To make the scenarios like the real
network, a network traffic generator called Ostinato is used to generate traffic into the network.

1.5 Scope of the Study

The scope of this study is concerned to Quality of Service Seamless MPLS Networks by applied
traffic engineering.

Page 4
1.7. Limitation of the Study
Due to memory limitations of personal computers and the process intensiveness of the simulation
tools used, it is not possible to power on more than 20 nodes (routers) simultaneously in
simulation environments and additional routers for redundancy and load balancing purposes are
used only in the core network domain. But it should be noted that increasing number of routers
for testing and analysis will not alter the overall result. I could not see the real result because it
was made by simulation.

1.8. Thesis Layout

There are four chapters in this thesis. The thesis is introduced in the first chapter. It contains
background information, a statement of the topic, the study's aims, the methods used to attain the
objectives, the thesis' scope and limitations, the thesis' contributions, and related works. Basic
ideas in MPLS technology, Seamless MPLS, traffic engineering, RSVP-TE signaling, and QoS
parameters are covered in Chapter 2. It emphasizes the benefits of MPLS over older
technologies, and the most frequent MPLS terms are briefly introduced in this chapter. The
network design is described in detail in Chapter 3. This comprises two scenarios: one is a
standard seamless MPLS network, and the other is RSVP-TE. The two scenarios' QOS
parameters were used to demonstrate and assess seamless MPLS. The simulation and result
analysis section discusses the simulation tools utilized, the simulation scenarios, the network
structure, and the analysis of the data collected.

Page 5
 CHAPTER TWO
REVIEW OF RELATED LITERATURE

2.1 Introduction to MPLS

MPLS is a telecommunications network forwarding mechanism that sends packets from one
node to the next based on short labels attached to packets rather than seeking up large IP
addresses at each router. This speeds up core routers that don't require sophisticated routing table
lookups. As a result, the data transfer speed is greatly increased [21].

Figure 2.1: Position of MPLS in OSI model

2.1.1 Forwarding Equivalence Class (FEC)

MPLS is a rating-based technology. When transmitting equivalence classes, it groups the packets
to be forwarded in the same way (FEC). Source address, destination address, source port,
destination port, kind of protocol, and VPN are some of the classification criteria that can be
used. The packets that belong to a specific FEC are then routed to the same symbol's Route
(LSP). When a packet comes, the router analyzes it to determine if it belongs to an existing
network. FEC is a logical object formed by the router, not a label or packet [35].

MPLs Label

A MPLS label is a short, fixed-length identifier pointing to a given FEC. A label may
reprehensibly only send a single FEC, but multiple labels may correspond to an FEC. The label
is part of the packet header and is only significant locally, as it carries no topology in formation
[43].

Page 6
 Figure 2.2: MPLS Packet Header (6)

2.1.2 Label Switching Router (LSR)

LSRs are the basic components of the MPLS network, and can be three different types depending
on their network function:

Label Edge Router (LER)-located on the network’s periphery and serves as a gateway between
the MPLS network and the WAN or the Internet. Could be an LER:
Ingress router – it is the entry point of the MPLS network. When a packet arrives it decides
whether the packet should be forwarded through the MPLS network, determines the FEC, the
packet belongs to, and encapsulates it with an MPLS header, based on the information it carries.
Egress router – it is the exit point of the MPLS network. It performs a normal IP look-up and
forwards the packet according to the appropriate IP routing protocol
Transit router – it is any router in the middle of the MPLS network and performs simple
switching, based on the label value
Penultimate router-it’s the router in the MPLS network before the last hop. The penultimate
router eliminates the MPLS header before forwarding the packet to the egress router, as the
packet will not be transferred to another transit router. The use of penultimate router
configuration is optional, since the egress router can also delete the MPLS header. The
penultimate router then operates

2.1.3 Label Switched Path (LSP)

From a particular FEC, the label switched path controls which direction packets will travel over
the MPLS network. The LSP is a one-way virtual circuit that connects the input to the egress
router. A signaling protocol, such as LDP or the Traffic Engineering Resource Reservation

Page 7
Protocol, is used to create the LSP (RSVPTE). Figure 2.3 shows a simplified diagram of an
MPLS network that demonstrates the LSP definition.

Figure 2.3: MPLS Network

2.1.4 Label Distribution Protocols (LDP)

The LDP protocol allows LSRs to request distribute and release binding information on the
mark. The label distribution method is called hop-by - hop forwarding and is chosen by the
underlying routing protocol of the Interior Gateway Protocol (IGP) along the normally routed
paths. The resulting LSPs are then used to relay label traffic over the MPLS network.

2.2 Label Distributions

When the packet enters the topology of the MPLS, LSR receives the packet and imposes it on the
packet via the Label Switch Path to the MPLS label and forward to the next hop. When the
packet reaches the next LSR, i.e. the intermediate LSR, the incoming label is swapped with the
outgoing label and the packet is transmitted. When the packet is received by the egress LSR it
strips off the packet label and forwards it to the destination router.
All LSRs in the MPLS network have IGPs (e.g., EIGRP, RIP, OSPF, etc.) running over the
network [7]. To order to accomplish the function of label delivery, neighboring LSRs need to
decide on the label used by each IGP prefix. The swapping of incoming and outgoing labels
should be visible for each LSR. Since the labels are local to neighboring routers and have no
global significance across the network, we need a mechanism to tell the routers which label to be
used when the packets are being forwarded. Two neighboring routers therefore need some kind

Page 8
of contact between them to decide on which mark to use for a given prefix. Otherwise the routers
get no idea of the packets being exchanged. The Label Distribution Protocol is required for this
purpose, or to complete the label distribution. Distribution of marks takes two separate ways [7]:
 Piggyback the labels on an existing IP routing protocol.
 Have a separate protocol distribute labels.

2.2.1 Piggyback the Labels on an Existing IP Routing Protocol

LSRs do not need new protocol in this method but they need to extend the existing routing
protocol to carry labels. This method has a great advantage because the routing and label
distribution are always synchronized which means that both labels and prefixes should be
present. Implementation for the distance vector routing protocol, e.g. EIGRP, which originates
the prefix from the routing table, is very simple.

2.2.2 Separate Protocol for Label Distribution

This label distribution method requires a separate protocol to distribute the labels and allows
distribution of the prefixes by the routing protocol. This approach has the advantage of routing
protocol independently and the downside is that each LSR needs a new protocol. There are
several protocol varieties which distribute labels including:

 Tag Distribution Protocol (TDP)

 Resource Reservation Protocol (RSVP)
 Label Distribution Protocol (LDP)

TDP was the first protocol developed and implemented by Cisco for label distribution. LDP was
later designed and developed by IETF. TDP and LDP operate in a similar way, but LDP has
more functionality than TDP. Due to the easy availability of LDP, TDP was replaced by LDP in
a very short time frame. RSVP is only used for MPLS traffic engineering.

2.3 Control Plane and Forwarding Plane

Control plane and forwarding plane are the part of router architecture. Control plane collects the
information that is used to forward the incoming packets. While forwarding plane decides how to
switch the incoming packets after being received at inbound interface [7].

Page 9
2.3.1 Control Plane
With the adjacent routers, the control plane exchanges routing information and labels. It is made
up of two types of protocols: routing protocols (e.g., RIP, EIGRP, OSPF, and BGP) and label
exchange information protocols (e.g., RIP, EIGRP, OSPF, and BGP) (e.g., LDP, TDP, RSVP,
etc.).

2.3.2 Data Plane

The forwarding plane of the data plane is based on the information connected to labels. Label
Information Base (LIB) and Label Forwarding Information Base (LFIB) are the two sorts of
tables (LFIB). The data plane uses LFIB to forward the tagged packets. The LIB table stores all
of the local labels as well as the mapping of the labels received from neighboring routers.

2.6. Seamless MPLS

The demand for a single converged packet network that can provide both fixed and mobile
networks, regardless of access type, grows on a regular basis. The performance of MPLS in core
networks and the benefits it provides have paved the way for its use in aggregation and access
networks as an alternative to ATM or traditional Ethernet-based aggregation. The widespread
deployment of mobile backhaul infrastructure has necessitated the integration of mobile
backhaul and core networks [9]. Deploying a service from one MPLS region to another
necessitates provisioning in the end-to-end network at numerous intermediate points, making
troubleshooting and fault recovery more difficult. A possible solution is deployment of
Architecture of one end-to - end infrastructure and transport network [3].

2.6.1 Deterministic End-to-end Service Restoration

Seamless MPLS is a robust network that offers an end-to - end deterministic service restores
(Sub-50ms), and there are two wide categories of functions that help to accomplish this. The first
set of functions includes ways to allow fast detection of performance degradation events and
fault location. The second set of functions consists of the required recovery activities necessary
for rerouting and restoring services. Failure Detection Mechanisms: There are various failure
detection mechanisms available. Layer2 failure detection relies on Ethernet Operation,
Administration, and Maintenance (OAM) capabilities, as well as integration of Bidirectional
Forwarding Detection (BFD) mechanisms with LSP and pseudo wires. For Layer3 fault

Page 10
detection and to test data plane consistency of pseudo wires, both single hop and multi-hop BFD
specified in RFC 5883 and RFC 5884 are supported for BGP sessions and targeted LDP sessions.

2.6.2 Decoupled Network and Service Architectures

Other end-to - end MPLS options (e.g., end-to - end LDP in a flat network) do not contain IGP or
MPLS signaling information within the region and are exchanged across regions. This increases
the size of the routing/ forwarding tables within individual routers, as well as the MPLS state.
Consequently, the Seamless MPLS architecture decouples the service and transport layer and
integrates access, aggregation and core into a single platform that supports residential, wholesale,
mobile, and business subscribers [3]. One of the major advantages is that problems on the
transport layer can to be solved once (and all providers have the solutions available). With
Seamless MPLS the use of service-specific configurations on intermediate nodes is not
necessary; all services can be deployed end-to-end [2].

2.6.3 Service Flexibility with Simplified Provisioning and Operations

Seamless MPLS architecture suggests a systematic way to enable end-to - end MPLS between
access nodes, with all MPLS-based forwarding labels. Using this method, packets are classified
at the entry point of the access network and are distributed all over the network to the receiving
end as classified packets. Therefore, service delivery and operations are significantly simplified,
reducing the number of provisioning points of service, and making the topological location of
service delivery points highly flexible.

2.6.4 Building Scalable Networks

Seamless MPLS helps to scale the end-to - end network to more than 100,000 MPLS devices,
understanding that certain nodes (e.g., access) have limited functionality, and are usually
designed for simplicity and lower cost. To deliver that scale, this clearly requires some new
thinking and innovative techniques.

2.7. Seamless MPLS Architecture

The handling of the total size of the necessary routes and MPLS label in the formation control
plane and forwarding plane state resulting from the specified scalability goals, in particular with
regard to the total number of access nodes, is one of the main elements to consider when

Page 11
designing architecture for a Seamless MPLS network. The intra-domain routing within each of the
MPLS domains (i.e. aggregation domains and core) utilize standard IGP protocols like OSPF or ISIS. A
systemic way to enable end-to - end MPLS across a single domain. This architecture has limited
provisioning flexibility since it is intimately tied with the topological arrangement of network nodes, and
it requires dealing with several technologies for troubleshooting and fault recovery on the operational
side.

Figure 2.4: MPLS with multiple domains [4]

With Seamless MPLS, the idea is to provision the service end-to-end and minimize the number
of provisioning points. The service provisioning is in-line with the network architecture,
maintains simplicity in the access network, and relies on increased capabilities and intelligence
on the service nodes. At the same time, it also simplifies operations and makes efficient use of
network resources by reducing the number of provisioning points and relying on a single MPLS-
based forwarding scheme in the data plane MPLS domains (regions) can be of different types:
IGP area, IGP instance or BGP AS, all spanned by a single MPLS network, with any to any
MPLS connectivity. Each area is in charge of connection (both IP and MPLS) within the region
and can choose whether to use LDP, RSVP-TE, or even LDP-over-RSVP on its own. Inter-
region connection is handled by region boundary nodes using an LSP hierarchy based on
"labeled BGP."
Since the service is initiated as an MPLS pseudo wire from the origination point at the access
node, any topological changes in the access can be easily made without having to completely re-
provision the service layer. This can be a significant operational asset to mobile backhaul access,
for example, where re-parenting of cell site routers to a different base station controller/radio
network controller (BSC/RNC) is a common occurrence.

Page 12
 Figure 2.5: Inter-AS seamless MPLS architecture

2.8. Traffic Engineering

The goal of traffic engineering is to improve the performance of operating networks. In general,
it refers to the use of technology and scientific principles to monitor, model, characterize, and
regulate Internet traffic, as well as the application of that knowledge and method to meet specific
performance goals. Measurement and control are two components of traffic engineering that are
relevant to MPLS. One of the main goals of Internet Traffic Engineering is to make network
operations more efficient and dependable while also maximizing network resource consumption
and traffic throughput.

2.8.1 Refresh Reduction Techniques

The Bundle message defined in RFC 2961 [3] packs a number of RSVP messages sent to the
same RSVP neighbor within a single larger RSVP message. To this purpose, a new RSVP
bundle message is defined: this message has its own header and a body which is made up with a
sequence of RSVP messages. Bundling is performed on a per-hop base.

2.8.2 RSVP with Traffic Engineering Extensions

Resource Reservation Protocol (RSVP) by its name we knows that it reserves the resource for
the network. The main task of the RSVP is that it reserves the bandwidth between the source and
destination along the defined path. In order to get the information in the network there is a router
in the network which will send the message packets in the network in order to get the details
regarding the bandwidth. There are four main messages which are used by the RSVP protocol
and these are given below:

1) RSVP PATH message

Page 13
 2) RSVP error message
3) RSVP RESERVATION message
4) RSVP tear message

2.8.3 RSVP PATH Message

In this message the head ended router is responsible for the reserve path. The main function of
the head ended router is that it sends the messages to the other routers and find out the path
information and send it to the head ended router. After that head ended router will decide that
which path is free from source to the destination?

2.8.4 RSVP RESERVATION Message

After the analysis of the Head ended router it will apply the resource reservation on the required
path and then use it in an effective manner. In the above figure when head ended router send
message to the inside router then it will send it to the next router and when it found that the
resource is free then it will generate the reservation message and finally it will come back to the
head ended router.

2.8.5 RSVP error Message

In this mechanism when head ended router send path message to the inside router there are
basically two conditions first one is the suppose the inside router don’t have any resources then it
will send back the message that I don’t have any resource right now with the identification of the
PATHERR Message. At the start there is no handling of the resource. On the other hand, if there
is a resource is available then it will send message to the other side most probably the destination
and suppose at destination side there is no resource are available then it sends an error message
then it will wait for some time according to the limitation when it gets any resource then it will
be successful otherwise it generates an error.

2.8.6 RSVP Tear Message

In case of RSVP tear message it will generate two types of tear messages one is for the clear of
the resources and the other one is for the clear of the path. When head ended router reserves any
resources then it will send back message that is called tear message.

Page 14
2.9. Quality of Service (QoS)

Quality of Service (QoS) is a broad word. It gives varying levels of treatment to various types of
traffic or applications that pass across the network. Different network requirements apply to
these applications. It must be the driving force behind various administrative policies that
regulate applications according to their specific needs. QoS within a network is essential to
guarantee the requirements of today’s converged networks. QoS provides that different levels of
service for business-critical application and delay-sensitive applications. QoS is to manage the
following network elements [14] [15].
 Bandwidth /Throughput
 Jitter and
 Delay
 Packet loss.

2.9.1.Through put
Throughput is a measure of how much information units a system can process within a given
amount of time. Given the dynamic nature of traffic flow across a network, differences in
resources at different times can become bottlenecks.

Table 2.1: Quality standards for throughput

Category Throughput/Bandwidth
Excellent 100%
Throughput Good 75%
standard Medium 50%
Poor <25%

2.9.2. Delay
RFC 7679 defines a metric for measuring one-way delay as the difference in the time at which
the datagram crosses two reference points. The delay of a datagram experienced within a service
provider network is defined as the difference in the time at which the datagram enters the
network and the time at which it leaves the network. It is also commonly referred to as latency.
Delay in TCP/IP networks can be classified as packetization delay, queuing delay, propagation
delay, transmission delay and processing delay [13].

Page 15
 Table 2.1: Quality standards ITU-TG114 for delay

Category Delay(MS)
Good 0-150
Delay standard Medium 150-400
Poor >400

2.9.3. Jitter
A metric for quantifying one-way jitter was defined in RFC 3393. Jitter is the variance in
network delay that datagrams undergo. Buffer space is made available in network nodes and
datagrams are buffered to avoid losing datagrams when a resource is temporarily congested.

Table 2.3: Quality standards ITU-T G 114 for Jitter

Category Jitter(ms)
Good 0-20ms
Jitter Standard
Medium 20-50ms
Poor 50>ms

2.9.4. Packet Loss

Traffic loss characterizes the datagram drops that occur in the path of a one-way traffic flow
between source and destination node. Having buffer space to temporarily queue datagrams in
network nodes helps reduce datagram loss, but it cannot be completely eliminated. Some of the
factors that contribute to datagram loss are [13] [14]:

 Congestion - Burstyn traffic can cause queue overflows resulting in datagram loss.
 Traffic rate limiting - In order to ensure customer traffic is conforming to a negotiated
SLA, service providers may rate-limit incoming traffic and drop nonconforming
datagrams.

Table 2.4: quality standard for packet lose

Category Packet loss

Excellent 0%
Packet loss standard Good 3%
Medium 15%
Poor 25%

Page 16
2.10. Review of Related Works
The authors in [7] have described the MPLS is considered as a routing method, and it is not a
facility or a service. MPLS can be encased with any prevailing infrastructures, namely digital
subscriber line, asynchronous transfer mode, frame relay, and IP. MPLS is not platform
dependent. It can work seamlessly without making any change in the current environment of
these technologies.

Quality of Service and the Network performance for selected traffic can be improved using the
MPLS. Different types of services bring in multiple classes of service. For example, an
enterprise’s mission-critical application can be placed in the top most section of service; the
applications less important might be in the Second best level of service, recreational applications
may be in the lower class services; separate class of service can be given to reduce the jitter for
VoIP traffic [2].

The researchers in [4] present a term application layer protocol refers to a protocol that manages
data and information exchanged between devices and the final software applications. A device
can either be a single sensor; a gateway that integrates a larger number of sensors; or even
actuating devices, such as motors, lighting equipment.

The Multiprotocol Label Switching (MPLS) network implemented at the AMIK Indonesia
College. After the model design is carried out, the network implementation is continued, so that
the MPLS network performance will be tested and compared to the performance without MPLS
using the model the research team planned [5].

Through IP/MPLS technology, [8] the seamless MPLS connects the access layer, convergence
layer, and backbone layer, and provides flexible and scalable networking architecture for
operators. It is improper to directly inherit all technologies from the old IP network.

After the devices of each layer are seamlessly connected, the scale of the IP/MPLS domain
improves by [9], orders of magnitude compared with the original networks. For example, in a
network with 20,000,000 users, if each DSLAM connects 100 users in FTTC access mode, the
number of nodes in the entire network is over 200,000. If each OLT connects 1000 users in
FTTB/FTTH access mode, the number of nodes in the entire network is 20,000.

Page 17
As shown in [6], QoS is mostly used for measuring various kinds of multimedia data. Sharing
(uploading) on SC has become the daily activity of end-users. As a result of this activity, it
provides an open challenge for service providers. As a service provider, the host delivers
productive infrastructure, allowing end-users to upload and share their high-quality images.

As shown in [11], in addition, many access devices, such as DSLAMs and OLTs, are available in
the network, taking up a high ratio of network investment. Hence, the introduction of the
IP/MPLS should not obviously affect the cost of access devices. In Seamless MPLS, networking,
the complexity of the access device control plane and performance specifications of the
forwarding layer must be reduced.

The internet engineering task force (IETF) standard in RFC 8277 [13] and RFC 7032 [14] are
aimed to address the drawbacks of traditional MPLS such as scalability and flexibility in service
provisioning limitations. The scalability is achieved by using label distribution protocol (LDP)
Downstream-on-Demand (DoD) label advertisements. MPLS traffic engineering, ATM, Layer 3
VPNs were configured, simulated and performance tested using the lab environment discussed in
the next section.

Benchmarking and testing performance testing details were based from RFCs 2544 and 5695. As
previously mentioned, pre-requisites such as dynamic routing protocols, network analysis, and
other concepts can be explored by collapsing or expanding on the lab network discussed [15].

In the original networking mode, the order of magnitude of the number of nodes in the backbone
and metro route domains is in the 1,000s. Hence, the scale of the route domain in Seamless
MPLS networking increases by an order of magnitude of one or two. In a large-scale network,
engineers must consider how to construct the route and MPLS tunnel, and how to guarantee the
availability of the networks [10].

The problem of how to extend QoS capabilities across multiple provider domains has not been
solved satisfactorily to date. The source of the problem lies mainly with the autonomous nature
of Internet Service Providers (ISP) and their loose federation that forms the global Internet [1].

Similarly routers can also identify traffic on known servers but old routers cannot difference
between traffic on higher level of protocols they are unable to differentiate between multiple http

Page 18
applications but new routers have solved this problem. QoS requirements for multimedia traffic
have been covered by different standardization groups, like ITU, ETSI or 3GPP [18].

As in [3], the Segment Routing is a promising Traffic Engineering (TE) model that provides end-
to-end communications SR can observably improve the network utilization and control the
routing path flexibly by encoding route information into a list of segments, i.e., the Segment List
(SL). The key feature of SR is that it adopts the source routing paradigm, which implies the
routing path followed by a packet is determined and written to the packet header by the first
switch of SR networks (called Ingress SR switch).

End points like video conferencing bridges can successfully mark the traffic but some time
network managers try to avoid endpoint classification is cases where the user may be able to
change the personal priority for online e-gaming [17].

The way to provide QoS in IP networks has been discussed for a long time [19]. The most
accepted solutions are IETF’s Internal Service and Difference Service both Internal Service and
Difference Service endow the routers with QoS mechanisms, such as queuing, scheduling and
shaping, as illustrated [20].

IP traffic can also manage voice and video data until less user traffic exists but as soon as the
traffic increases through user request the packets travelling the same IP destination path become
lost or slow due to OSPF congestion. So the quality of service guarantee voice and video data is
no more accomplished. There is no standard way to provide QoS to voice and video data packets
in IP packet transmission [16].

The internet engineering task force (IETF) standard in RFC 8277 [13] and RFC 7032 [14] are
aimed to address the drawbacks of traditional MPLS such as scalability and flexibility in service
provisioning limitations. The scalability is achieved by using label distribution protocol (LDP)
Downstream-On-Demand (DoD) label advertisements. To enhance the flexibility in
provisioning, label mapping information for a route is piggybacked in the same BGP update
message that is used to distribute the route itself.

Page 19
2.10.1 Review of Related Works Summary
Table 1.1: Summary of Review of Related Works
Author Improved QoS Technique/Approach Tool(s) Used Metrics Gap(s) Identified
Feature(s) Used Used
Asrat MB et End-to-end weighted fair Huawei’s delay, jitter, Chassis
al., [47] QoS queueing (WFQ) for Enterprise packet loss clustering of
parameters of congestion Network and traffic access &
MP-BGP management and Simulation utilization aggregation
MPLS VPN of weighted random Platform devices, using
EthioTelecom early detection (eNSP) and LDP to label
service level (WRED) for Wireshark MPLS down
agreement congestion avoidance streaming.
(SLA)
customers.

Yalemzewd Impact of Buffers packets in Emulated Delay, SR-domain with

MB [24] Segment queues upon network Virtual Jitter, LDP only
Routing MPLS congestion and use a Environment Packet loss capable routers
on end-to-end scheduling algorithm -Next and Packet on endto-end
QoS parameter to determine the Generation sequencing. QoS as most of
of forwarding order. (EVE-NG) the router
SR-MPLS Monitors network especially in the
QoS. resource usage and access domain
drops packets to may not be
mitigate network SRMPLS
overload if capable.
congestion worsens.
Heng Luo Mobile ad hoc Routing protocols GloMoSim, Delay, jitter Not multiple
[37] network DSDV, a typical OPNET, and access techniques
(MANET), proactive protocol QualNet and throughput. such as Bluetooth
Best Effort and DSR, a typical MATLAB and MANETs.
QoS Support reactive protocol and
Routing in rank them
accordingly.
Yalemzewd end-to-end Nonstop active
Enterprise Throughput, Border Gateway
Negash [40] network QoS routing (NSR)-
Network latency, Protocol (BGP)
performance enabled control plane
Simulation packet loss does not support
by classical protocols. loop-free
Platform and jitter QoS
MPLS. alternate (LFA)
(eNSP), implementation
support for ISIS,
Network across multiple
OSPF and LDP. Quality domains (inter-
Analyzer domain or inter-
(NQA), AS).
Ostinato
Omair MPLS over Technique for traffic Cisco Packet Throughput, Using MPLS that
Ahmad and ATM method engineering and Tracer queuing not helped to
Shakeel of media functioning of delay and replace Frame
Ahmed[2] traffic routing. multiple services and Bandwidth Relay, dedicated
it adds up the leased lines and

Page 20
 benefits like having offers a new
greater scalability option for WAN
that can improve connectivity.
network operations.
Fathurrahmad MPLS VPN MPLS works on GNS3, Bandwidth That may not
and Salman networks and packages with MPLS Microsoft and produce realistic
Yusuf [5] providing a headers, traffic Visio throughput. network
stable network engineering Network performance
bandwidth processes and Design 2016 related results
efficiency and implement a VPN and Virtual
is used at network Machines.
AMIK.
Nokia Evolving to LSPs, by using a MATLAB Maximum Information such
Ahmed [34] end-to-end technique such as the and CPLEX. Link as measured
MPLS gradient method Utilization delay and
architectures Internet traffic rate, Packet residual
loss rate bandwidth. It’s
and high complexity
Demand
Rejection
rate.
M. Tanvir Decreasing A Differentiated NS2 Packet loss, MPLS QoS
and A. Packet Loss Service (DiffServ) simulator Packet mechanism also
Said[45] for QoS based approach is delay. cannot be
Sensitive IP proposed for QoS considered to
Traffic in provisioning. improve the
DiffServ performance of
Enabled the traffic.
Network Using
MPLS TE.
Cortese Creation and QoS from end to end MATLAB packet loss These aggregate
Gandr [26] Deployment of between network and models is that
End-User elements. throughput they mostly
Services in capture steady
Premium IP state behavior
Networks. because the
averaging is
typically done
over large time
scales
J.barakovic, Multimedia Traffic engineering MATLAB Packet loss In this paper
H. Bajric and traffic analysis and multi service and CPLEX. and Packet importance of
A.Husic [21] of MPLS and functionality. delay. MPLS and the
Non MPLS. need to
implement it in
order to
overcome the
limitations
provided by
ATM.

Page 21
 CHAPTER THREE
SIMULATION RESULTS AND ANALYSIS

The purpose of this chapter is to give a quick overview of the simulation approaches used to
study service quality. The simulation scenarios and network topologies utilized in the analysis
are listed below. At the conclusion, the simulation experiments and their interpretation are
discussed.

3.1 Overview of Simulation Tools

3.1.1 Graphical Network Simulator-3 (GNS3)

GNS3 is an emulator for network software that was initially released in 2008. It enables the use
of both virtual and physical devices to mimic complicated networks [20]. Cisco IOS is simulated
using Dynamics emulation software. [20] Many significant corporations, like Exxon, Walmart,
AT&T, and NASA, employ GNS3, which is also popular for network professional certification
exam preparation. The software has been downloaded 11 million times as of 2015.+[26] In this
paper, a network architecture with XRv routers as edge routers and IOSv routers as core routers
is implemented in a GNS3 environment.

3.1.2 VMware Workstation

It allows you to run numerous operating systems on a single Windows or Linux computer. Create
real Linux and Windows virtual machines as well as other desktop, server, and tablet
environments, complete with configurable virtual networking and network condition simulation,
for use in code development, solution architecture, application testing, and product
demonstrations, among other things.

3.1.3 GNS3 VM
The GNS3 VM is a requirement if you want to run Qemu based devices on Windows or Mac OS.
It is a virtual machine that you import into VMware Workstation (recommended) or Virtual Box
on your local PC when running a local version of the GNS3 VM. It can also be used in a

Page 22
distributed environment where you run the client software (GNS3-all-in-one) on your local PC
and the GNS3 VM on a Hypervisor such as ESXi or the cloud.

3.1.4 Cisco Internetworking Operating System (IOS) images

Cisco routers and modern Cisco Network Switches use the IOS operating system software. The
image file for Cisco IOS (Internetwork Operating System) is usually stored in flash memory and
has a naming convention.

3.1.5 Ostinato
Ostinato is a user-friendly packet maker, network traffic generator, and analyzer. Create and
deliver packets using multiple protocols at varied rates for several streams. Ethernet/802.3/LLC
SNAP; ARP, IPv4, IPv6, IP-in-IP, IP Tunneling (6over4, 4over6, 4over6, 4over6, 4over6,
4over6, 4over6, 4over6, 4over6, 4over6, 4over6, 4over6, 4over6, 4over6, 4over6, 4over6,
4over6, 4over6, 4over6, 4over6, 4over6, 4over6, 4over6, 4over6, 4over6, 4over6, 4over6,
4over6, 4 Client-server architecture is also supported. It has the ability to build and configure
sequential and interleaved streams of various protocols at various rates. A user-defined script
[18] also allows for the addition of any unimplemented protocol.

3.1.6 Paessler Router Traffic Grapher (PRTG)

PRTG is a network monitoring tool that helps you to ensure that your computer systems are
running smoothly and that no outages occur.

3.2 Simulation Scenarios and Network Topology

In the implementation phase, the enterprise network simulation framework Graphical Network
Simulator-3 is used to establish a realistic environment, and two scenarios with various
parameters are built to assess the performance of the Seamless MPLS and RSVP-TE Seamless
MPLS. Ostinato is utilized in both situations to generate network traffic and analyze network
efficiency using four QoS indicators. Finally, using PRTG network monitoring tools, test results
for the two scenarios are retrieved from the simulator. Seamless MPLS architecture (scenario 1)
and RSVP-TE Seamless MPLS architecture (scenario 2) are depicted in Figures 3.1 and 3.2,
respectively. In both scenarios three MPLS domains are used where access and aggregation
regions are in one MPLS domain and the core network is in another MPLS domain

Page 23
 Figure 3.1: Seamless MPLS architecture

Figure 3.2: RSVP-TE Seamless MPLS architecture

3.3 Seamless MPLS Network topology details

Table 3.1: Topology details

Number of router 10
Number of links 14
Router model 7200
Router operating system 7200 Software (C7200-ADVENTERPRISEK9-
M), Version 15.2(4)S3,
Cloud interface Network monitoring (PRTG)
Ostinato Traffic generator

3.4 Router interface configuration

Page 24
 Table 3.2: Router interface details

10.10.10.1/32 is directly connected, Loopback0

10.10.10.9/32 is directly connected, Tunnel0
R1 10.10.12.0/30 is directly connected, GigabitEthernet2/0
10.10.12.1/32 is directly connected, GigabitEthernet2/0
10.10.13.0/30 is directly connected, GigabitEthernet1/0
10.10.10.2/32 is directly connected, Loopback0
10.10.12.0/30 is directly connected, GigabitEthernet2/0
R2 10.10.12.2/32 is directly connected, GigabitEthernet2/0
10.10.24.0/30 is directly connected, GigabitEthernet1/0
10.10.24.1/32 is directly connected, GigabitEthernet1/0
10.10.10.3/32 is directly connected, Loopback0
10.10.13.0/30 is directly connected, GigabitEthernet1/0
10.10.13.2/32 is directly connected, GigabitEthernet1/0
10.10.34.0/30 is directly connected, GigabitEthernet3/0
R3 10.10.34.1/32 is directly connected, GigabitEthernet3/0
10.10.34.4/30 is directly connected, GigabitEthernet3/0.34
10.10.34.5/32 is directly connected, GigabitEthernet3/0.34
10.10.35.0/30 is directly connected, GigabitEthernet2/0
10.10.35.1/32 is directly connected, GigabitEthernet2/0
10.10.10.4/32 is directly connected, Loopback0
10.10.24.0/30 is directly connected, GigabitEthernet1/0
10.10.24.2/32 is directly connected, GigabitEthernet1/0
10.10.34.0/30 is directly connected, GigabitEthernet3/0
R4 10.10.34.2/32 is directly connected, GigabitEthernet3/0
10.10.34.4/30 is directly connected, GigabitEthernet3/0.34
10.10.34.6/32 is directly connected, GigabitEthernet3/0.34
10.10.46.0/30 is directly connected, GigabitEthernet2/0
10.10.46.1/32 is directly connected, GigabitEthernet2/0
10.10.10.5/32 is directly connected, Loopback0
10.10.35.0/30 is directly connected, GigabitEthernet2/0
10.10.35.2/32 is directly connected, GigabitEthernet2/0
R5 10.10.56.0/30 is directly connected, GigabitEthernet1/0
10.10.56.1/32 is directly connected, GigabitEthernet1/0
10.10.57.0/30 is directly connected, GigabitEthernet3/0
10.10.57.1/32 is directly connected, GigabitEthernet3/0

10.10.10.6/32 is directly connected, Loopback0

10.10.46.0/30 is directly connected, GigabitEthernet2/0
10.10.46.2/32 is directly connected, GigabitEthernet2/0
R6 10.10.56.0/30 is directly connected, GigabitEthernet1/0
10.10.56.2/32 is directly connected, GigabitEthernet1/0
10.10.68.0/30 is directly connected, GigabitEthernet3/0
10.10.68.1/32 is directly connected, GigabitEthernet3/0
10.10.10.7/32 is directly connected, Loopback0
10.10.57.0/30 is directly connected, GigabitEthernet3/0
10.10.57.2/32 is directly connected, GigabitEthernet3/0
10.10.78.0/30 is directly connected, GigabitEthernet2/0

Page 25
R7 10.10.78.1/32 is directly connected, GigabitEthernet2/0
10.10.78.4/30 is directly connected, GigabitEthernet2/0.78
10.10.78.5/32 is directly connected, GigabitEthernet2/0.78
10.10.79.0/30 is directly connected, GigabitEthernet1/0
10.10.79.1/32 is directly connected, GigabitEthernet1/0
10.10.10.8/32 is directly connected, Loopback0
10.10.68.0/30 is directly connected, GigabitEthernet3/0
10.10.68.2/32 is directly connected, GigabitEthernet3/0
10.10.78.0/30 is directly connected, GigabitEthernet2/0
R8 10.10.78.2/32 is directly connected, GigabitEthernet2/0
10.10.78.4/30 is directly connected, GigabitEthernet2/0.78
10.10.78.6/32 is directly connected, GigabitEthernet2/0.78
10.10.81.0/30 is directly connected, GigabitEthernet1/0
10.10.81.1/32 is directly connected, GigabitEthernet1/0
10.10.10.9/32 is directly connected, Loopback0
10.10.79.0/30 is directly connected, GigabitEthernet1/0
R9 10.10.79.2/32 is directly connected, GigabitEthernet1/0
10.10.91.0/30 is directly connected, GigabitEthernet2/0
10.10.91.1/32 is directly connected, GigabitEthernet2/0
10.10.10.10/32 is directly connected, Loopback0
10.10.81.0/30 is directly connected, GigabitEthernet1/0
R10 10.10.81.2/32 is directly connected, GigabitEthernet1/0
10.10.91.0/30 is directly connected, GigabitEthernet2/0
10.10.91.2/32 is directly connected, GigabitEthernet2/0

3.5 Router configuration for IP network

Each router in an IP network must be configured with routing protocols in order to advertise its
network and to determine the path to the target address.
IP network configuration
Table 3.3: IP network details
Routing protocols OSPF
Router R1 up to R10

3.6 Router configuration for Seamless MPLS network

Tag switching must be enabled on all interfaces of a router that belongs to the MPLS domain in
order to participate in an MPLS network. It's also a good idea to enable the protocol that
publishes label numbers across the network (similar to how IP networks advertise subnets).
RSVP makes advantage of the TE possibilities whereas LDP employs the MPLS forwarding
technique. Tables 3.4 and 3.5 demonstrate the MPLS network configuration details.

Page 26
MPLS network configuration

Table 3.4: MPLS network details

Routing protocols Multi process number OSPF
Label distribution protocol LDP

Table 3.5: Seamless MPLS-TE tunnel details

Tunnel number Source Destination
tunnel0 10.10.10.1 10.10.10.9
tunnel0 10.10.10.9 10.10.10.1
Label distribution protocol RSVP-TE

Table 3.6: Unified MPLS components

Routing protocol and label distribution Three BGP domains

BGP 3107 R1 to R3 ,R3 to R7 and R7 to R9

3.7 Simulation Parameters Analysis

3.7.1 Throughput Analysis

The amount of data sent or received by a network or entity, or the amount of data processed in a
certain time period, is referred to as throughput. Bits per second (bit/s or bps) are the basic units
of measurement. Due to system losses and delays, the throughput may be lower than the input
tax. Throughput is a good indicator of a communications link's channel capacity. To calculate the
throughput, we must first determine how much data is being communicated and how long it
takes to complete the data transmission. For both situations (Seamless MPLS (SMPLS) and
RSVP TE-SMPLS), a traffic generator is employed to produce traffic at a rate of 1Mbps into the
network. During the simulation, PRTG collects relevant information including traffic in, traffic
out and total traffic. For different file sizes the test results for both scenarios are tabulated in the
Table 3.7 shown below.

Page 27
 Table 3.7: Output of Throughput for Seamless MPLS and RSVP TE-Seamless MPLS at
different file sizes
File size(bytes) 100 200 300 400 500 600 800 1100
Average RTT (Sec) for Seamless- MPLS 10.07 11.93 14.3 15.3 16.61 17.32 23.03 28.81
Average RTT (Sec) for RSVP-TE Seamless- MPLS 6.8 8.34 8.8 9.75 12.15 14.46 18.37 24.05
Throughput Seamless-MPLS (Kbps) 79.44 134.1 167.8 209.15 240.8 277.13 277.9 305.4
Throughput RSVP-TE Seamless-MPLS (Kbps) 117.6 191.8 272.7 328.2 329.2 331.95 348.4 365.9

Throughput
400
Throughput(kbit/s)

300
200
100
0
0 200 400 600 800 1000 1200
File size Kbyte/s

Throughput Seamless-MPLS (Kbps) Throughput RSVP-TE Seamless-MPLS (Kbps)

Figure 3.1: Graph of throughput for scenarios 1 and 2

Figure 3.2: PRTG Graph of throughput for scenarios 1 and 2

Figure 3.3 displays a graph of simulation throughput results. At small file sizes, there is no
notable difference in performance between the two scenarios, but as file sizes grow larger,

Page 28
 RSVP-TE Seamless MPLS outperforms Seamless MPLS. For example, if we compare file sizes
of 1100Kbytes, the throughput difference is approximately 104.9Kbps, or 19.8%. There are
several reasons why RSVP-TE Seamless MPLS has a higher throughput than Seamless MPLS.

3.7.2 Latency Analysis

The ITU-T recommendation for maximum end-to-end latency is 150ms. Although the Same
topology setup as in the throughput analysis is used for latency analysis, The ICMP (Internet
Control Message Protocol) ping test type is used to send test probe. To collect the average
completion time of each test probe at different data sizes as shown in Table 3.8. Network traffic
is generated end-to-end traffic and PRTG collects the minimum, maximum and average delay of
sending test message. For the simulation three congestion levels are considered: data rate less
than link bandwidth, data rate equal to link bandwidth and data rate less than link bandwidth.

Table 3.8: Output of latency for Seamless MPLS and RSVP TE-Seamless MPLS at
different file sizes

File size(bytes) 100 200 300 400 500 600 800 1100
Ave Completion Time(ms) for Seamless MPLS 58 64 75 85 94 106 117 127
Ave Completion Time(ms) for RSVP-TE 54 56 68 79 81 94 102 110
Seamless MPLS
Latency different 2 8 7 6 13 12 15 17
Latency different in % 4 14 10.29 7.59 16.04 12.76 14.7 15.45

Latency
150
Latency(MS)

100

50

0
0 200 400 600 800 1000 1200
File Size (byte)

Ave Completion Time(ms) for Seamless MPLS

Ave Completion Time(ms) for RSVP-TE Seamless MPLS

Page 29
 Figure 3.3: Graph of latency for scenarios 1 and 2

The latency versus file size graph in Figure 3.5 shows that the Seamless MPLS (Scenario 1) has
higher latency than Seamless RSVP-TE MPLS (Scenario 2). On average the latency difference
between the two scenarios is about 21.4%. That is the latency of Seamless RSVP-TE MPLS is
improved on average by 10s (i.e. 21.45%) compared to the Seamless MPLS counterpart.

3.7.3 Packet Loss Analysis

When one or more packets of data being transmitted through the internet or a computer network
fail to reach their destination, this is known as packet loss. Signal degradation, excessive loads
on network lines, malformed packets being deleted, and network element defects can all cause
packet loss. The recommended value for packet loss according to ITU standards is less than 3%.
Congestion is chosen as a source of packet loss in this study, which means that the links in the
network are purposely congested by utilizing network traffic generators to bring additional
network traffic into the network. ICMP test type is used to send test probes at three conditions i.e
data rate less than link bandwidth, data rate equal to link bandwidth and data rate greater than
link bandwidth.

Table 3.9: Output of packet loss for Seamless MPLS and RSVP-TE Seamless MPLS at
different file sizes

File size(bytes) 20 1000 2000 3000 4000 5000 6000 8000

Packet Loss for Seamless MPLS 0 0 0 0 1 4 12 22
Packet Loss for RSVP-TE Seamless MPLS 0 0 0 0 0 2 8 13
Packet Loss different 0 0 0 0 1 2 4 18
Packet Loss in % 0 0 0 0 1 1

Page 30
 Packet Loss
25
packet loss %

20
15
10
5
0
0 1000 2000 3000 4000 5000 6000 7000 8000 9000
File size(Byte)

Packet Loss for Seamless MPLS Packet Loss for RSVP-TE Seamless MPLS

Figure 3.4: Graph of packet loss for scenarios 1 and 2

As shown in Table Figure 3.6 the other method by which RSVP-TE Seamless MPLS minimizes
the effects of congestion and link failure is by using fast reroute and pre-computed alternative
routes. These mechanisms help to use alternative paths and nodes in a sub-second to reduce
packet loss. The performance of RSVP-TE Seamless MPLS is much better than Seamless MPLS

3.7.4 Jitter Analysis

Jitter is the delay variation and is introduced by the variable transmission of delay of the packets
over the network. This can occur because of routers' internal queues behavior in certain
circumstances (e.g. flow congestion), routing changes, etc. This parameter can seriously affect
the quality of streaming audio and/or video. To handle jitter, it is needed to collect packets and
hold them long enough until the slowest packets arrive in time, rearranging them to be played in
the correct sequence. Jitter buffers can be observing when using video or audio streaming
websites (e.g. YouTube) and are used to counter jitter introduced by the internet so that a
continuous play out of the media transmitted over the network can be possible. When clicking in
a link to play the video, buffering starts before the media stream actually does. This procedure
causes additional delay, but is necessary in the case of jitter sensitive applications. We can
simulate and calculate the jitter values using the same procedures and scenarios as for the other

Page 31
parameters. Network congestion is a common jitter factor and our simulation considered it
primarily for jitter analysis.

Table 3.10: Output of jitter for Seamless MPLS and RSVP-TE Seamless MPLS at different
file sizes
File size(bytes) 36 1000 2000 3000 4000
Jitter (ms) for Seamless MPLS 10 14 16 17 19
Jitter (ms) for RSVP-TE Seamless MPLS 7 10 12 14 16

Jitter
40
Jitee ms

20

0
0 1000 2000 3000 4000 5000 6000 7000 8000 9000
Data size byte

Jitter (ms) for sMPLS Jitter (ms) for RSVP-TE sMPLS

Figure 3.5: Graph of Jitter for scenarios 1 and 2

As shown in Figure 3.7, the simulation results of average jitter for RSVP-TE Seamless MPLS is
smaller than that of Seamless MPLS. Take data size of 8000 bytes for example, the jitter
difference is about 3ms (i.e. 12.5%). This performance difference between the two scenarios
have significant impact on jitter sensitive real time traffic such as voice, video conference, live
streaming, etc.

Page 32
3.8. Comparative Analysis of Results
Author Improved Technique/App Tool(s) Metrics Gap(s) Compare
QoS roach Used Used Used Identified
Feature(s)
Yalemzew Impact of Buffers packets Emulated Delay, SR-domain Quality of
d MB [24] Segment in queues upon Virtual Jitter, with LDP service of
Routing network Environme Packet loss only capable seamless
MPLS on congestion and nt -Next and Packet routers on MPLS by
end-to-end use a Generation sequencing endto-end applying
QoS scheduling (EVE-NG) . QoS as most resource
parameter of algorithm to of the router reservation
SR-MPLS determine the especially in protocol
QoS. forwarding the access (RSVP) End-
order. Monitors domain may to-end users.
network not be
resource usage SRMPLS
and drops capable.
packets to
mitigate
network
overload if
congestion
worsens.
Heng Luo Mobile ad Routing GloMoSim Delay, Not multiple Techniques
[37] hoc network protocols , OPNET, jitter and access through TE,
(MANET), DSDV, a QualNet throughput. techniques Seamless
Best Effort typical and such as MPLS
QoS Support proactive MATLAB Bluetooth limitations with
Routing in protocol and and influence on
DSR, a typical MANETs. TE
reactive
protocol and
rank them
accordingly.
Yalemzew end-to-end Nonstop active Enterprise Throughput Border Support QoS
d Negash network QoS routing (NSR)- Network , latency, Gateway implementation
[40] performance enabled control Simulation packet lossProtocol with MPLS.
by classical plane Platform and jitter (BGP) does
MPLS. protocols. loop- (eNSP), not support
free alternate Network QoS
(LFA) support Quality implementati
for ISIS, OSPF Analyzer on across
and LDP. (NQA), multiple
Ostinato domains
(inter-domain
or inter-AS).
Omair MPLS over Technique for Cisco Throughput Using MPLS Evolved
Ahmad ATM method traffic Packet , queuing that not through ATM
and of media engineering Tracer delay and helped to and frame relay
Shakeel traffic and functioning Bandwidth replace VAN networks;

Page 33
Ahmed[2] routing. of multiple Frame Relay, MPLS uses
services and it dedicated labels to
adds up the leased lines advertise
benefits like and offers a between
having greater new option different
scalability that for WAN routers by
can improve connectivity.means of label
network mapping
operations. through label
switching
mechanism
Fathurrah MPLS VPN MPLS works GNS3, Bandwidth That may not The resulting
mad and networks and on packages Microsoft and produce
LSPs are then
Salman providing a with MPLS Visio throughput. realistic
Yusuf [5] stable headers, traffic Network network used to relay
network engineering Design performance
label traffic
bandwidth processes and 2016 and related
efficiency implement a Virtual results over the MPLS
and is used at VPN network Machines.
network.
AMIK.

This Work Improve the QoS services in Graphical Bandwidth, Independent MPLS by
QoS of MPLS Network
jitter, delay network applying
Seamless implementation Simulator- traffic
MPLS and 3 (GNS3, and packet treatment (RSVP) End-
networks by communication Mware for
loss. to-end users.
applying techniques Workstatio
customer
resource through TE. n and Techniques
oriented GNS3 VM VPN and through TE,
traffic
engineering traffic Seamless
techniques.
classificatio MPLS.
n.

Page 34
 CHAPTER FOUR
CONCULUSIONS AND FUTRURE WORKS

4.1 CONCLUSIONS
In this article, the quality of service seamless multiprotocol level switching is compared to two
network scenarios, one with RSVP-TE seamless MPLS and the other without. We employed four
KPIs of QOS characteristics to analyze and compare Seamless MPLS with RSVP-TE Seamless
MPLS: throughput, latency, packet loss, and jitter. To analyze the lap report, we use GNS3 to
establish the two topologies with the needed configuration files, Ostinato to produce network
traffic, PRTG network monitoring tools to collect data, and PRTG and Excel to present the
results. RSVP-TE allows user packets to be tunneled inside an RSVP LSP to an LDP far-end
destination (with the benefits of RSVP LSPs, fast-reroute (FRR), and traffic engineering (TE))
and does not follow the IGP, making separate judgments on reserved traffic.

This functionality is primarily used to implement MPLS-based services, such as VPRN, VLL,
and VPLS, in big networks where a full mesh of LSPs has reached its scalability constraints. In
general, we find that the RSVP-TE Seamless MPLS was more reliable than seamless MPLS
based on the analyses and results obtained.

4.2 CONTRIBUTIONS
Operators of MPLS networks, such as Ethio telecom, must optimize their existing network
infrastructure, expand the options accessible to existing services, and maybe generate new
service offerings without investing more resources. Segment Routing (SR) provides the
necessary capability to meet these requirements. In addition to MPLS core networks, many
service providers, like Ethio telecom, have introduced mobile backhaul. The interconnection, as
well as service delivery, between these multiple domains, should be smooth, with no additional
delay, signaling protocol overhead, or flexibility issues, ensuring the best end-to-end QoS and
resource usage. Using new emerging technologies, this thesis intends to improve end-to-end
network QoS performance by optimizing the Unified MPLS, which is based on SR architecture.
This reduces the limits of traditional MPLS architecture and improves service delivery
scalability, resource usage, and flexibility in any telecommunications business. For
organizations, the study will provide a better knowledge of concepts for information about

Page 35
MPLS importance, usage, and deployment. MPLS is a novel technology for designing and
implementing QoS services and application classes that are reliable, secure, efficient, and
standard. For traffic engineering and multicasting, this technique will provide long-term
solutions. Internet service providers, satellite operators, and network providers are all involved in
the investigation of IP and MPLS in satellite Internet.

This thesis aims to improve end-to-end network QoS performance by using newly emerging
technologies to optimize the traditional MPLS architecture. This reduces the limits of traditional
MPLS architecture while also increasing the scalability and flexibility of service delivery in any
telecommunications industry. It provides new ways of thinking and methodologies to networking
systems research in general. The researcher have tried to look at my participation MPLS and
simulation MPLS, But then given the time we have and also have seen things in science and tried
to apply what we can able. Although the researcher did not make a big contribution to science by
doing this, we knew what the study needed to know.

4.3 FUTURE WORKS

While the thesis has accomplished all of the objectives set forth in Chapter One, there are still
some concerns to be resolved.

 The following are the inquiries: Study and implement a seamless MPLS solution in
Ethiopian telecommunications networks. It is first necessary to identify the different
types of routers (or devices) that are utilized in an end-to-end IP network. MPLS and
Seamless MPLS capability must be checked on the access devices.
 For independent network traffic treatment, analyze the service per customer VPN and
traffic classification.

Page 36
References
[1]. D. Griffin, J. Grien, J. Spencer, P. Georgatsos, and P. Morand, “Inter domain routing through QoS-class
planes,” in Communication Magazine, vol. 45, no. 2, pp. 88-95, Feb.2007.
[2]. A. Shakeel and O. Ahmad, “Performance Comparison of MPLS and ATM Based Networks,” Selected Areas
in Communications, February. 2021.
[3]. K. Alemayehu, “Analyzing Impact of Segment Routing MPLS on QoS,” in Communication Magazine,
December.2019.
[4]. C. Akasiadis, V. Pitsilis and D. Spyropoulos, “A Multi-Protocol IoT Platform Based on Open-Source
Frameworks,” September. 2019.
[5]. Fathurrahmad, S. Yusuf, T. Iqbal, A. Salam, “Virtual Private Network (VPN) Network Design For
Multiprotocol Label Switching (MPLS),” November. 2019.
[6]. S. Karim, H. He, A. Ali , H. Magsi & A.Laghari, “Quality of service (QoS): measurements of image
formats in social cloud computing,” September 2020.
[7]. S. Rajagopalan, “A Study on MPLS Vs SD-WAN,” June. 2021.
[8]. L. Xiao, J. Wang, K. Lui, and K. Nahrstedt, “Advertising inter-domain QoS routing information,” IEEE
Journal on Selected Areas in Communications, vol. 22, pp. 1949 – 1964, Dec. 2004.
[9]. Resilient Service Enablement at Massive Scale Using Standard Protocols, Juniper Networks, 2012
[10]. D. Adami, "A new ns2 module for the simulation of MPLS networks with point-to-multipoint LSPs
support,” IEEE International Conference on Communications (ICC 2009), Dresden, Germany, June 2009
[11]. Z. Li, “Inter-SDN (SDNi) in Seamless MPLS for Mobile Backhaul Network,” draft li-rtgwg-sdni-seamless-
mpls-mbh-00 (work in progress), March 2017.
[12]. S. Mishra, S. Sonavane, and A. Gupta, “Study of traffic generation tools,” International Journal of Advanced
Research in Computer and Communication Engineering (IJARCCE), vol. 4, no. 6, pp. 159–162, Jun. 2015.
[13]. E. Rosen, "Using BGP to Bind MPLS Labels to Address Prefixes," RFC 8277, DOI 10.17487/RFC8277,
October 2017.
[14]. M. Tanvir and A. Said, "Decreasing Packet Loss for QoS Sensitive IP Traffic in DiffServ Enabled Network
Using MPLS TE," in International Symposium in Information Technology, Kuala Lumpur, Malaysia, 2010.
[15]. Tony Li, Procket Networks Inc., MPLS and the Evolving Internet, IEEE Communications Magazine,
December 1999.
[16]. Davie, B., Y. Rekhter, A. Viswanathan, S. Blake, V. Srinivasan, and E. Rosen, Use of Label Switching With
RSVP, draft-ietf-mpls-rsvp-00.txt, March 1998.
[17]. Callon, R., A. Viswanathan, and E. Rosen, Multiprotocol Label Switching Architecture, draft-ietf-mpls-
arch-02.txt, July 1998.
[18]. J. M. Chung, “Analysis of MPLS traffic engineering,” Proceedings of the IEEE Midwest Symposium on
Circuits and Systems, San Francisco, CA,USA, August 2000, pp. 550–553.
[19]. E. Osborne and A. Simha, Traffic Engineering with MPLS, Indianapolis, IN, Cisco Press, 2003.

Page 37
[20]. A. Bott et.al., "A tool for the generation of realistic network workload for emerging networking scenarios,”
Computer etworks (Elsevier), vol. 56, no. 15, pp 3531-3547, 2012.
[21]. J. barakovic, H. Bajric and A.Husic, “Multimedia traffic analysis of MPLS and Non MPLS”, June 2006.
[22]. M.Hafeez, Asif, M.Golam, “Performance comparison of IP, MPLS and ATM based networks Cores using
OPNET”, 2006.
[23]. MD. A. Rahman, A. H. Kabir, K. A. M. Lutfullah, M. Z. Hassan, M. R. Amin “Performance Analysis and
the Study of the behaviour of MPLS Protocols”, May 2008.
[24]. Yalemzewd Negash, “Analyzing Impact of Segment Routing MPLS on QoS’’ , December 2019.
[25]. A. Cuevas et al., “Usability and Evaluation of a Deployed 4G Network Prototype”, Journal of
Communications and Networks (ISSN: 1229-2370), Vol. 7, No. 2, pp. 222-230, June 2005.
[26]. G. Cortese et al., “CADENUS: Creation and Deployment of End-User Services in Premium IP Networks”,
IEEE Communication Magazine, Vol. 41, No. 1, pp. 54-60, January 2003.
[27]. Labovitz, A. Ahuja, R. Wattenhfore, and S. Venkatachar, “The impact of internet policy and topology on
delayed routing convergence,” in Twentieth Annual Joint Conference of the IEEE Computer and
Communications Societies, Proc. IEEE, vol. 1, 2001, pp. 537-546.
[28]. Dr. A. S. Ahmad, Dr. T. Alatky, and M. Jafar, “Performance Analysis DiffServ based Quality of Service in
MPLS Networks,” International Journal of Scientific & Engineering Research, vol. 6, no. 9, Sep. 2015.
[29]. B. Quoitin, S. Uhlig, C. Pelsser, L. Swinnen, and O. Bonaventure, “Inter-domain traffic engineering with
BGP,” IEEE Communications Magazine Internet Technology Series, vol. 41, no.5, May 2003.
[30]. B. Davie and Y. Rekhter, MPLS Technology [2]and Applications. Morgan Kaufmann Publishers, 2000.
[31]. N. Kumar and G. Saraph,” End-to-End QoS in Inter-Domain Routing,” in ICNS ’06: Proceedings of the
International conference on Networking and Services, IEEE Computer Society, 2006, pp. 82
[32]. T. G. Griffin, F. B. Shepherd, and G. Wilfong, “The stable path problem and inter-domain routing,”
IEEE/ACM Trans. Networking., vol. 10, pp. 232-243, Apr. 2002.
[33]. N. Leymann, “Seamless MPLS Architecture,” draft-leymann-mpls-seamless-mpls-07(work in progress),
June 2014.
[34]. Nokia, “Evolving to end-to-end MPLS architectures,” Finland, White paper, SR1610000967EN, Oct. 2016.
Analysing the impact of Seamless MPLS on QoS Page 62.
[35]. Juniper Networks, Inc., “Building Multi-Generation Scalable Networks with End-to-End MPLS,” CA 94089
USA, White paper, 2000452-001-EN, Jan. 2012.
[36]. D. Griffin, J. Grien, J. Spencer, P. Georgatsos, and P. Morand, “Interdomain routing through QoS-class
planes,”in Communication Magazine, vol. 45, no. 2 [5], pp. 88-95, Feb.2007.
[37]. L. Xiao, J. Wang, K. Lui, and K. Nahrstedt, “A Best Effort QoS Support Routing in Mobile ad hoc
Networks,” Dec. 2018.
[38]. M. Boucadair and P. Morand, “A solution for providing inter-as mpls-based qos tunnels,” Draft-boucadair-
pce-interas-01.txt (work in progress), October 2005.

Page 38
[39]. R. Atkinson, E. S. Floyd, and Internet Architecture Board, "IAB Concerns and Recommendations Regarding
Internet Research and Evolution," RFC 3869, DOI 10.17487/RFC3869, August 2004,
[40]. Yalemzewd Negash,” Analysing Impact of Seamless MPLS on QoS”, August 2018,
[41]. T. Griffin and B. Presmore, "An Experimental Analysis of BGP Convergence Time," in Proc. IEEE ICNP,
2001.
[42]. C. Labovitz, A. Ahuja, A. Bose, and F. Jahanian, "Delayed Internet Routing Convergence," in Proc. ACM
SIGCOMM, 2000.
[43]. T. Bu, L. Gao, and D. Towsley, "On Routing Table Growth," in Proc. IEEE Global Internet Symp., 2002.
[44]. O. Nordstrom and C. Dovrolis, "Beware of BGP attacks," SIGCOMM Computer Communications Review,
vol. 34, no. 2, pp. 1-8, Apr. 2004.
[45]. M. Tanvir and A. Said, "Decreasing Packet Loss for QoS Sensitive IP Traffic in DiffServ Enabled Network
Using MPLS TE," in International Symposium in Information Technology, Kuala Lumpur, Malaysia, 2010,
pp. 789-793.
[46]. B. Miller and E. Stewart, “Multi-Protocol Label Switching (MPLS) Conformance and Performance
Testing,” Ixia, White paper, Jan. 2004.
[47]. Beyene, Asrat Mulatu, and Shimelis Asrat Argaw. "Improving Quality of Service of Border Gateway
Protocol Multiprotocol Label Switching Virtual Private Network of Ethio Telecom Service Level
Agreements." In International Conference on Information and Communication Technology for Development
for Africa, pp. 278-288. Springer, Cham, 2019.

Page 39
 APPENDIXES

Appendix: Scripts for Seamless MPLS Configuration and RSVP-TE

Seamless MPLS Configuration
PE_1#show running-config
Building configuration...
Current configuration : 3291 bytes
!
version 15.2
service timestamps debug datetime msec
service timestamps log datetime msec
!
hostname PE_1
!
boot-start-marker
boot-end-marker
!
vrf definition VPN1
rd 65108:20
!
address-family ipv4
route-target export 65108:10
route-target import 65108:10
exit-address-family
!
no aaa new-model
no ip icmp rate-limit unreachable
!

Page 40
no ip domain lookup
ip cef
no ipv6 cef
!
mpls traffic-eng tunnels
!
mpls traffic-eng path-option list name R3-R4-R6-R8-R7-R9
path-option 1 explicit identifier 1
multilink bundle-name authenticated
!
ip tcp synwait-time 5
!
interface Loopback0
ip address 10.10.10.1 255.255.255.255
!
interface Tunnel0
ip unnumbered Loopback0
tunnel mode mpls traffic-eng
tunnel destination 10.10.10.9
tunnel mpls traffic-eng path-option 1 explicit name TO_R4
tunnel mpls traffic-eng record-route
!
interface FastEthernet0/0
no ip address
shut down
speed auto
duplex auto

Page 41
!
interface FastEthernet0/1
no ip address
shut down
speed auto
duplex auto
!
interface GigabitEthernet1/0
description TO.....R3
ip address 10.10.13.1 255.255.255.252
ip ospf network point-to-point
negotiation auto
mpls ip
mpls label protocol ldp
mpls traffic-eng tunnels
!
interface GigabitEthernet2/0
description TO....PE_2(R2)
ip address 10.10.12.1 255.255.255.252
ip ospf network point-to-point
ip ospf cost 10
negotiation auto
mpls ip
mpls label protocol ldp
mpls traffic-eng tunnels
!
interface GigabitEthernet3/0
vrf forwarding VPN1

Page 42
ip address 10.130.200.1 255.255.255.252
negotiation auto
!
interface GigabitEthernet4/0
no ip address
shutdown
negotiation auto
!
interface GigabitEthernet5/0
no ip address
shut down
negotiation auto
!
interface GigabitEthernet6/0
no ip address
shut down
negotiation auto
!
router ospf 1
router-id 10.10.10.1
passive-interface Loopback0
network 10.10.0.0 0.0.255.255 area 0
mpls traffic-eng router-id Loopback0
mpls traffic-eng area 0
!
router bgp 65108
bgp log-neighbor-changes
neighbor 10.10.10.3 remote-as 65108

Page 43
neighbor 10.10.10.3 update-source Loopback0
neighbor 10.10.10.4 remote-as 65108
neighbor 10.10.10.4 update-source Loopback0
neighbor 10.10.10.9 remote-as 65108
neighbor 10.10.10.9 update-source Loopback0
!
address-family ipv4
network 10.10.10.1 mask 255.255.255.255
neighbor 10.10.10.3 activate
neighbor 10.10.10.3 send-label
neighbor 10.10.10.4 activate
no neighbor 10.10.10.9 activate
exit-address-family
!
address-family vpnv4
neighbor 10.10.10.9 activate
neighbor 10.10.10.9 send-community extended
exit-address-family
!
address-family ipv4 vrf VPN1
redistribute connected
redistribute static
default-information originate
exit-address-family
!
ip forward-protocol nd
!
!

Page 44
no ip http server
no ip http secure-server
ip route 10.10.10.9 255.255.255.255 Tunnel0
ip route vrf VPN1 0.0.0.0 0.0.0.0 10.130.200.2
!
ip explicit-path name TO_R4 enable
next-address loose 10.10.10.4
next-address loose 10.10.10.8
!
route-map TO_TE permit 10
set interface Tunnel0
!
mpls ldp router-id Loopback0
!
control-plane
!
line con 0
exec-timeout 0 0
privilege level 15
logging synchronous
stopbits 1
line aux 0
exec-timeout 0 0
privilege level 15
logging synchronous
stopbits 1
line vty 0 4
login

Page 45
end
PE_2#show running-config
Building configuration...
Current configuration : 2018 bytes
!
version 15.2
service timestamps debug datetime msec
service timestamps log datetime msec
!
hostname PE_2
!
boot-start-marker
boot-end-marker
!
no aaa new-model
no ip icmp rate-limit unreachable
!
no ip domain lookup
ip cef
no ipv6 cef
!
mpls traffic-eng tunnels
multilink bundle-name authenticated
!
ip tcp synwait-time 5
!
interface Loopback0
ip address 10.10.10.2 255.255.255.255

Page 46
!
interface FastEthernet0/0
no ip address
shut down
speed auto
duplex auto
!
interface FastEthernet0/1
no ip address
shut down
speed auto
duplex auto
!
interface GigabitEthernet1/0
description TO.....R4
ip address 10.10.24.1 255.255.255.252
ip ospf network point-to-point
negotiation auto
mpls ip
mpls label protocol ldp
mpls traffic-eng tunnels
!
interface GigabitEthernet2/0
description TO......PE1(R1)
ip address 10.10.12.2 255.255.255.252
ip ospf network point-to-point
ip ospf cost 10
negotiation auto

Page 47
mpls ip
mpls label protocol ldp
mpls traffic-eng tunnels
!
interface GigabitEthernet3/0
no ip address
shut down
negotiation auto
!
interface GigabitEthernet4/0
no ip address
shut down
negotiation auto
!
interface GigabitEthernet5/0
no ip address
shut down
negotiation auto
!
interface GigabitEthernet6/0
no ip address
shut down
negotiation auto
!
router ospf 1
router-id 10.10.10.2
passive-interface Loopback0
network 10.10.0.0 0.0.255.255 area 0

Page 48
mpls traffic-eng router-id Loopback0
mpls traffic-eng area 0
!
router bgp 65108
bgp log-neighbor-changes
neighbor 10.10.10.3 remote-as 65108
neighbor 10.10.10.3 update-source Loopback0
neighbor 10.10.10.4 remote-as 65108
neighbor 10.10.10.4 update-source Loopback0
!
ip forward-protocol nd
!
no ip http server
no ip http secure-server
!
mpls ldp router-id Loopback0
!
control-plane
!
line con 0
exec-timeout 0 0
privilege level 15
logging synchronous
stopbits 1
line aux 0
exec-timeout 0 0
privilege level 15
logging synchronous

Page 49
stopbits 1
line vty 0 4
login
end

Page 50