OPNET Modeler and Ns-2: Comparing the Accuracy Of Network
Simulators for Packet-Level Analysis using a Network Testbed
Gilberto Flores Lucio, Marcos Paredes-Farrera, Emmanuel Jammeh, Martin Fleury, Martin J. Reed
Electronic Systems Engineering Department
University of Essex
Colchester, Essex C04 3SQ
United Kingdom
gflore@essex.ac.uk, http://privatewww.essex.ac.uk/~gflore/
Abstract: - This paper presents a comparative study of two well-known network simulators: OPNET Modeler and Ns-
2. Other studies in this area have generally been confined to just one simulator. The motivation of this paper is to
provide a guide to researchers undertaking packet-level network simulations. The simulator outputs were compared to
the output from a live network testbed. The experimental comparison consisted of deploying both CBR data traffic,
and an FTP session, both on the network testbed and the simulators. CBR data traffic was used due to its simplicity of
modeling. A custom analysis tool was employed to examine the behavior of the network in different background
traffic scenarios. The same scenarios were then recreated in the simulators in order to ascertain their realism. The
results show the necessity of fine-tuning the parameters within a simulator so that it closely tracks the behavior of a
real network.
Keywords: network simulator, OPNET Modeler, Ns-2, simulation methodology
1 Introduction
Network simulators have grown in maturity since
they first appeared as performance, management and
prediction tools. Simulators are normally used as
network management tools, for which packet level
analysis is not commonly employed. However, more
studies are needed to establish guidelines for
researchers so that they may select and customise a
simulator to suite fine-grained packet level analysis
[1][2]. Reference [3] reports `the breach of credibility'
that studies based on simulation tools need to tackle;
one motivation behind this paper is to address this
need. The ease and facility that simulators provide in
evaluating “radical” changes to a network environment
cannot be discarded. There are a considerable number
of simulations tools in the market. The main
characteristics that divide them are: accuracy, speed,
ease of use, and monetary expense.
This paper concentrates on the accuracy of the
simulation in comparison to a real network for packet
level analysis. Two of the currently popular network
simulators that can perform this type of analysis are
Modeler from OPNET
1
[4] and Ns-2 from the Virtual
Internetwork Testbed project VINT [5]. These are
selected because of their popularity within academia,
commercial and industrial communities [6].
An ideal way to test simulator accuracy is to measure
real network traffic and compare it with the simulators
1
OPNET Modeler was provided under the OPNET
University Programs
results. Therefore, this paper presents real and
simulated results (from the two simulator tools), where
CBR (Constant Bit Rate) data traffic, and an FTP (File
Transfer Protocol) session are generated in order to
test the network performance with different types of
traffic. CBR traffic was selected due to the simplicity
of its nature and FTP due to its popular use and
dynamic behavior. CBR in particular is a useful
analysis as it provides “ground truth” i.e. there are few
external factors that influence the model. Whereas
FTP was chosen at this stage as any more complex
protocol (e.g. HTTP) would have too many degrees of
freedom. The analysis is made at a packet-by-packet
level concentrating on the bandwidth utilisation
analysis.
The initial tests used the default configuration
profiles provided with the simulators, however, several
manual modifications were made at different stages of
the experiments to investigate possible improvements.
An in-house analysis tool called tcpflw was used to
examine the behavior of the real network. tcpflw
has successfully been used to analyze the performance
of a video streaming session [7]. The results reported
here were directly compared with the results from the
simulators.
This paper does not aim to define which network
simulator is best as there are too many different
parameter variations and different possible network
scenarios to adequately determine this in a single
paper. Instead, this paper demonstrates how the
suitability of simulators can be validated for the
particular case of packet level forwarding in an IP