You are on page 1of 4

1

MPLS vs. IP Network from A QoS Perspective

Mosaab Shawgi Omer Abdalla


Dept. Electrical and Electronic Engineering
University of Khartoum Faculty of Engineering
Khartoum, Sudan
Mosaab1288@hotmail.com

Abstract
optimizing
the
performance
of
telecommunication networks is one of the main traffic
engineering goals [1] by dynamically analyzing, predict the
network behavior and regulating of data transmitted over that
telecommunication network.
Multiprotocol Label Switching (MPLS) improve network
performance am reliable delivery of the Internet services with
high transmission speed and lower delays. The main features of
MPLS is its Traffic Engineering (TE) which is used for
effectively managing the networks in their efficient and
utilization of network resources [1]. In this thesis the
performance of Voice over Internet Protocol (VoIP) was
compared by simulating MPLS network and Internet Protocol
(IP) network. OPNET modeler 14.5 was used as a simulation
environment and both networks compared based on the quality
of service parameters Voice jitter, Voice packet Throughput,
voice delay variation.
Keywords: Optimized Network Engineering Tool
(OPNET), Multiprotocol Label Switching (MPLS), Traffic
Engineering (TE) and Voice over Internet Protocol (VoIP)
INTRODUCTION
Increased number of users online through internet made the
popular services such as media and broadcasting uses the
Internet as a medium to reach their customers. Providing the
Real time applications on Internet is a challenging task for the
conventional IP networks uses the best effort services which
doesnt provides guarantee of quality of services and traffic
engineering (TE) [2]. Multi-Protocol Label Switching (MPLS)
is a developed technology which has an important role in the
next Network generation by providing efficiency and increased
Quality of service (QoS) and Traffic Engineering (TE)[3].
Limitations like excessive delays and high packet loss of IP
networks it overcomes by providing scalability and congestion
control capability. Due to the low latency and low packet loss
during routing of packets MPLS is considered as good
environment for VoIP applications [4].
Many related works found and a collection was selected
those with the idea of comparative analysis of MPLS over

Dr. Amin Babiker A/Nabi Mustafa


Dept. Electrical and Electronic Engineering
University of Neelain Faculty of Engineering
Khartoum, Sudan
Amin31766@gmail.com

Non-MPLS networks and shows MPLS have an excellent


performance compared to traditional IP networks. In this paper
an evaluation to the MPLS and IP Network study was made for
MPLS on signaling protocols (CR-LDP, RSVP and RSVP-TE)
[5]for Traffic Engineering by discussing their functionality and
classification. Simulation of MPLS and Non-MPLS network
was completed and performance is compared by considering
the parameters such as system throughput, delay variation and
jitter on the network traffic. Qual-Net 4.0 simulator is used for
simulation purpose [6]. In second paper the author focus on the
analytical models to evaluate the efficiency of voice over IP
network with applications on MPLS network.[7] In this paper
network parameters are presented to support the quality of
service (QoS) requirements and traffic engineering standards
that is supported by MPLS.[8] A mathematical expressions was
used to evaluate the models for both IP and MPLS networks. In
other paper the objective was calculating the minimum
Number of VOIP calls and maximum number of VoIP calls
that can be established in an enterprise IP network.
PROBLEM DEFINITION
Network traffic congestion is the main issue that worldwide
is focusing into, many techniques, protocols, algorithms were
developed to help improve the network performance in term of
the QoS parameters.
In this work the planning and simulating the network
including traditional IP network compared to other techniques
such as MPLS which provide many algorithms help optimize
the performance was needed.
OBJECTIVES
The main goal in this thesis is to evaluate the performance
of two networks with a variety of traffic configurations in order
to evaluate the performance of enterprise IP networks compare
to MPLS network in terms of quality of services parameters
(QoS).

2
METHODOLOGY
1st reviewing literature is done from studying the case
studies and ethnographies and 2nd focus the state of problem
and how it can be solved. A simulation and analysis was
performed using OPNET Modeler 14.5 in order to get
statistical results or data. By analyzing the results a solution
that is provided for the problem statement of the thesis. For this
purpose a network model was designed and results are obtained
in order to compare the performance of VOIP application
within two network structures MPLS with IP network.
SIMULATION SCENARIOS
The Simulation was divided in two scenarios. Scenario1: In
this part of the simulation the VoIP traffic is send from source
part (VOIP_WEST) to the destination part (VOIP_EAST) in
the two networks the MPLS and IP networks. The main
Scenario is to compare the performance of VoIP traffic in the
both networks by using performance metrics, i.e., voice jitter,
packet delay variation, and throughput. The simulation results
obtained are analyzed to determine the efficient technology
used for transmitting VoIP traffic. Scenario 2: In this part, an
approach is made to estimate the approximate minimum
number of calls that can be maintained in the both networks.
This is done by designing the real network in the OPNET.

found in the graphs it is noticed that there is an increasing in


the performance when the Voice over IP (VoIP) traffic is
transmitted using MPLS Network Structure. For each scenario
the duration of the simulation is 15 minutes The VoIP traffic
starts at the 2.5 minutes and ends at the 15th minutes of the
simulation time. In both scenarios VoIP calls are added at fixed
time intervals i.e., for every two minutes starting from 2.5th
minutes till 15th minutes except the Voice packet jitter it starts
from 100th second and ends to 420 second.
Note: Red Sign represent the MPLS Network and
Network

Blue Sign represent IP

THROUGHPUT OF MPLS VS. IP NETWORK

MPLS SIMULATION MODEL


Figure (3) Throughput of Voice packet send and received
VOICE PACKET JITTER

FIGURE (1) MPLS SIMULATION SCENARIO


IP NETWORK SIMULATION MODEL

Figure (4) Voice packet jitter

FIGURE (2) IP SIMULATION SCENARIO


RESULTS AND ANALYSIS
The results shown in the following figures are the performance
metrics obtained for MPLS and IP networks. From the results

3
VOICE PACKET DELAY VARIATION

increase at 1st 240 sec in IP network and for MPLS network it


starts to increase at 300 second. The voice packet delay
variation shown in Figure 5 has same variations in graphs,
starts from the 2.00 minutes and increases to the end of
simulation time in discrete wayand it was found that the delay
variation of the MPLS networks has an increase delay
compared with the IP network and thats because the
algorithms that may use to detect equal cost path or to the
RSVP (Resource Reservation Protocol).
CONCLUSION

Figure (5) Voice packet delay variation


DISCUSSION
The limitations of IP networks when used to transfer voice
applications, we have identified some crucial factors why IP
networks are unsuitable for implementing VoIP applications.
IP networks offer low predictability of services and exhibits
high packet loss and delays which are unacceptable for the
real-time applications.
The performance of voice over MPLS compared with
Traditional IP results show that the performance of VoIP
application is increased when it is implemented in MPLS
network with considering TE. OPNET modeler 14.5 is used for
simulation. The performance metrics obtained from the
simulation shows that due to the added features like TE and
MPLS mechanism of forwarding packets make MPLS a better
choice in transferring the VoIP communication.
In figure 3 gives the average number of packets send and
received in both MPLS and conventional IP networks. By the
end of simulation it is observed that MPLS model gives more
throughput than the IP model.
The simulations of the MPLS and IP models are done
considering the background traffic. It is observed from the
figure1 that voice packets start to drop from 2.5 minutes in the
IP network whereas in MPLS voice packets are started to drop
from 3.5 minutes. In the simulation the early packet drop in IP
network indicates that it cannot establish the VoIP calls with
acceptable quality after 2.5 minutes. The VoIP calls established
after 3.5 minutes experiences loss of calls due to the packet
loss which cause call disconnect and voice skips.
The Voice packet drop in MPLS network starts at 3.5 minutes
due to the fact that MPLS delivery the packets with high
transmission speed with low delays and moreover the TE is
implemented in the MPLS network which temporarily reduces
the congestion. Due to these factors the packet drop in MPLS
networks starts at 3.5 minutes where as in IP network the
packet drop starts at 2.5 minutes, this increases the throughput
in the MPLS network.
The Figure 4 shows the Voice packet jitter of MPLS and IP
network model. It was found that the Voice Jitter starts to

The comparative analysis of MPLS network and


conventional IP network is made on focusing on the
performance metrics such as Voice jitter, Voice packet delay
variation, and Throughput. Based on the simulation results it
can be concluded that MPLS provides good environment for
VoIP application compared to enterprise IP networks because
of the following reasons
MPLS takes less processing time in forwarding the packets
which is more suitable for time critical application such as
VoIP
Implementing of MPLS with TE minimizes the congestion in
the network
MPLS has a minimum delay and provide a high throughput
compared to enterprise IP network
REFERENCE
[1] Mahesh Kr. Porwal., Anjulata Yadav., S. V. Charhate,
Traffic Analysis of MPLS and Non MPLS Network including
MPLS Signaling Protocols and Traffic distribution in OSPF
and MPLS, International Conference on Emerging Trends in
Engineering and Technology, ICETET, July 2008.
[2] Nader F.Mir., Albert Chien, Simulation of Voice over
MPLS communications Networks, IEEE ICSS02,conference.
[3] Rahman M.A.,Kabir A.H., Lutfullah K.A.M., Hassan M.Z.
and Amin M.R., Performance analysis and the study of the
behavior of MPLS protocols, International Conference on
Computer and Communication Engineering ICCCE, 13-15
May 2008, Page(s):226 229
[4] Haris Hodzic, Sladjana Traffic Engineering with
Constraint Based Routing Zoric 50th International
Symposium ELMAR-2008, 10-12 September 2008, Zadar,
Croatia.
[5] Junaid Ahmed Zubairi Ph.D., Member, IEEE Voice
Transport Techniques over MPLS
[6] Xipeng Xiao, Alan Hannan, and Brook Bailey, Global
Center Inc. Lionel M, NI, Michigan State University. Traffic
Engineering with MPLS in the Internet
[7] David Wright, University of Ottawa Voice over MPLS
Compared to Voice over Other Packet Transport
Technologies IEEE Communications Magazine November
2002.
[8] Md. Rahman ,Kabir A.H., Lutfullah K.A.M., Hassan M.Z.
and Amin M.R., Performance analysis of MPLS Protocols
over conventional,
.