Networking And Data Communications Computer Science Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Abstract - The basic scope of this coursework is to simulate and analyze the operation of a hospital's department network, using OPNET IT Guru Academic software simulation tool, under two main scenarios. The worst case scenario (heavy traffic) and the department's future expansion scenario. During this process the network topology of the hospital's department is presented, user profiles and supported applications are defined and, finally, the simulation results of each scenario are evaluated with respect to applications response times and link's utilization.


This paper introduces a way to design and evaluate a hospital's network through the OPNET simulation program that offers the ability to configure the parts of a topology and observe its behavior. The hospital's department uses the applications of Database Access, E-mail, Web Browsing, Digital Image View and Instant Messaging.

To begin with, the Opnet tool's operation and significance is discussed and a presentation and evaluation of the network's operation and performance takes place. The simulation will take place under medium load for all applications. Then, by setting the network's applications at the highest traffic load, a worst case scenario is created and the network is re-evaluated. Finally, regarding a future upgrade of the system, proposals will be presented and implemented to maintain the application response times to acceptable levels and then tested, by running the simulation one last time.

By quoting each application's response time graphs, as well as the link's utilization graphs, conclusions are easily drawn about the performance and efficiancy of each scenario, thus proving that simulation programs, such as OPNET, are precious tools for a network's design, development and maintenance.


Opnet is a very large and powerful software with a wide variety of possibilities to simulate an entire heterogenous network with various protocols. It was developed in 1986 by MIL3 Inc., nowadays known as OPNET Technologies Inc. The software was originally developed for the needs of military, but it has grown to be a world leading commercial network simulation tool. Opnet is quite expensive for commercial usage, but there are also free licenses for educational puproses, like the one on this paper.

Opnet is a high level event based network level simulation tool that operates at "packet-level". It was originally built for the simulation of fixed networks, since it contains a huge library of accurate models of commercially available fixed network hardware and protocols. Nowadays, the possibilities for wireless network simulations are, also, very wide. Opnet can be used as a research tool or as a network design and analysis tool. The threshold for the usage is high for the developer, but low for the end user.

Opnet consists of high level user interface, which is constructed from C and C++ source code blocks with a huge library of Opnet specific functions. Hierarchical structure is divided to three main domains: (i) The Network Domain (networks and subnetworks, network topologies, geographical coordinates, mobility), (ii) The Node Domain (single network nodes such as routers, workstations, mobile devices etc.) and (iii) The Process Domain (single modules and source code inside networn nodes, such as data traffic source model).

The most important tools of Opnet are the Network model editor, the Node model editor, the Process model editor, the Packet format editor, the Analysis configuration tool and the Simulation tool. This paper is going to be focused on the Node model editor and the Simulation tool.


Using OPNET Modeler the network will be designed step by step. The components inside the brackets next to each element reported below, express the exact hardware drawn from the Object Palette of the program. The two hospital department subnets are located in Athens, Greece. The ATH1 subnet is in the area of Piraeus and consists of the two radiology department LAN's, while the ATH2 subnet is located in the area of Marousi and contains the data centre of the enterprise. The two subnets are connected via a 1GbE link (1000BaseX):

Figure 1. Hospital Topology

The ATH1 subnet consists of the Radiology Department, located on the Floor 1 of Piraeus building and the Enterprise's Router, located at the Basement of the building. The Radiology Department consists of the Administrative_Personnel 100BaseT_LAN and the radiology Doctors 100BaseT_LAN, each including 10 workstations connected via 100MbE (100BaseT) links to a switch (ethernet16_switch), which, in turn, is connected via 100MbE link (100BaseT) to the general Radiology_Department_Switch (ethernet16_switch). Finally, the Radiology_Department_Switch (ethernet16_switch) is connected via a 100MbE (100BaseT) link to the Enterprise_Router (fr4_ethernet2_gtwy) located at the Basement of the Piraeus building:

Figure 2. ATH1 Subnet

The ATH2 subnet, located on the Floor 0 of the Marousi building, is the enterprise's Data Centre and consists of the Database_Server (ethernet_server), the Email_Server (ethernet_server), the Web_Server (ethernet_server), the X-Ray_Server (ethernet_server) and the IM_Server (ethernet_server), each connected via 100MbE (100BaseT) links to the Data_Centre_Switch (ethernet16_switch), which in turn is connected via a 100MbE (100BaseT) link to the Data_Centre_Router (fr4_ethernet2_gtwy):

Figure 3. ATH2 Subnet

First, the Application Configuration's attributes are edited and the 6 applications (6 rows) are added in the Applications Definitions, as shown below. The applications are the Database Query (information retrieval from database), the Email (upload and download), the HTTP (web page searching), the X-Ray Image View (download and view of patients' digital x-ray images), the Instant Messaging Login (login to the instant messaging service) and the Instant Messaging Exchange (exchange of messages between clients):

Figure 4. System's Applications

The descriptions of the Database, Email, Web, and IM_Login applications are easily set from the available descriptions of the program. Their Description as well as their Value for the default scenario are shown below:

Figure 5. Applications Description

The X-Ray and IM_Exchange applications need to be customized, by setting Task Descriptions to them.

Therefore, the next step is to edit the Task Configuration's attributes by adding 2 tasks (2 rows) in the Task Specification section, that describe the process of viewing the digital x-ray images and the exchange of instant messages:

Figure 6. System's Tasks

Then, each of these tasks is manually configured as shown below. The X-Ray_Image_Download Task consists of 1 phase, in which the client sends a request to server, which responds with a packet of 3.6MB as instructed by the ICT Manager.

Figure 7. Configuration of X-Ray_Image_Download Task

The IM_Exchange Task consists of 1 phase, the Message_Exchange phase where the client sends a request packet with an Inter-request time of 2.6sec, Mean Outcome of 205bytes and Variance of 20 and then receives a packet of the same characteristics from the server:

Figure 8. Configuration of IM_Exchange Task

It has to be noted, that all phases of the above Tasks have the following Transport Connection settings:

Figure 9. Transport Connection settings for all phases

So, back to the Application Configuration's settings, the tasks created above are added to the Description of the X-Ray and IM_Exchange applications:

Figure 10. Addition of X-Ray_App to the Application_configuration

Figure 11. Addition of IM_Exchange_App to the Application_configuration

Then, the Profile Configuration's attributes are edited and 2 profiles (2 rows) are added in the Profile Configuration section. One for the Administrative Personnel and one for the radiology Doctors. The Operation Mode for all three of them is set to Simultaneous, which means that they can run at the same time and not in series:

Figure 12. Network's Profiles

Then, applications are added for each profile, as shown below:

Figure 13. Administrative_Personnel's Profile Applications

Figure 14. Doctors' Profile Applications

The Repeatability, for every application of both profiles, will have these settings:

Figure 15. Profiles' Repeatability Settings

Then, on ATH2 subnet the attributes of each server are edited, setting as Server Address the name that has been given to each server node, but also defining which application each server will run, at the Application: Supported Services section. The Database_Server will run the Database Application, the Email_Server will run the Email Application, the Web_Server will run the Web Application, the X-Ray_Server will run the custom X-Ray Application and the IM_Server wll run the IM_Login Application and the custom IM_Exchange Application. The IM_Server settings are shown as an example below:

Figure 16. IM_Server settings

Figure 17. IM_server Supported Applications

Finally, the attributes of the Administrative_Personnel LAN and Doctors LAN must be edited. LAN Server Name must be the name that has been set to the node. Both LAN's have 10 Workstations:

Figure 18. Administrative_Personnel LAN and Doctors LAN main attributes

Also, the Application: Source and Destination Preferences for each LAN, as well as the Appications and Profiles, each LAN supports, have been set as shown below:

Figure 19. Administrative_Personnel LAN Settings

Figure 20. Doctors LAN Settings

The Actual Names of all the Symbolic servers in the Destination Preferences for both LAN's, are the actual servers. For example HTTP Server is Web_Server and IM Server is IM_Server.


Now, that every part of the network system has been designed and set, the network's performance must be assessed by means of suitably-chosen statistics. These statistics are the response times of each application, shown below; the Custom Application (X-Ray application and IM application), the DB Query (Database application), the Email (Email application) and the HTTP (Web application):

Figure 21. Chosen Application Statistics

Also, the statistics of Utilization on the 1GbE line that interconnects the two subnets, is chosen:

Figure 22. Chosen Line Statistics

The simulation of this default scenario is, then, run and the following results, with respect to response time of each application, and utilization of the 1GbE link are viewed in average mode:

Figure 23. Database application graph

Figure 24. Email (Download) application graph

Figure 25. Email (Upload) application graph

Figure 26. Web application graph

Figure 27. IM Login application graph

Figure 28. IM Exchange (Administrative_Personnel) application graph

Figure 29. IM Exchange (Doctors) application graph

Figure 30. X-Ray application graph

Figure 31. Link Utilization graph


Judging from the above graphs, it is clear that the network is operating satisfactory both by means of the application response times and the utilization of the main link that connects the two subnets.


In order to test the network's performance at maximum traffic load, a worst case scenario must be implemented, where all applications will be in High/Heavy Load. The new Values of the applications' Descriptions are shown below:

Figure 29. Worst Case Scenario Chosen application statistics

After run the simulation of the worst case scenario, the results compared to the default scenario ones are expressed in the following graphs:

Figure 30. Database default VS worst case graph

Figure 31. Email Download default VS worst case graph

Figure 32. Email Upload default VS worst case graph

Figure 33. Web default VS worst case graph

Figure 34. IM Login default VS worst case graph

Figure 35. IM Exchange (Administrative_Personnel) default VS worst case graph

Figure 36. IM Exchange (Doctors) default VS worst case graph

Figure 37. X-Ray default VS worst case graph

Figure 38. Link Utilization wost case graph


Judging from the above graphs, it is clear that in the worst case scenario there is an increase in all the applications' response time, except for the two custom ones, the X-Ray and the IM, where the response times remain almost the same.


In this section, the future expansion of the network is examined and cost efficient solutions are being proposed, in order for the network to work sufficiently under heavy load and even under new additional load. At first, all the servers' processors are being upgraded to triple-core processors with five fold speeds each. The new settings of the IM Server is shown below as an example:

Figure 39. Upgraded IM Server

Next, all the links of the network are upgraded to 1GbE (1000BaseX) links, excpept, of course, for the main link that is already of that bandwidth. Having made these upgrades, the simulation is run one last time and the results compared to both previous scenarios are being shown below (Blue: default scenario, Red: worst case scenario, Green ):

Figure 40. Database with upgrade graph

Figure 41. Email Download with upgrade graph

Figure 42. Email Upload with upgrade graph

Figure 43. Web with upgrade graph

Figure 44. IM Login with upgrade graph

Figure 46. IM Exchange (Administrative_Personnel) with upgrade graph

Figure 45. IM Exchange (Doctors) with upgrade graph

Figure 47. X-Ray with upgrade graph

Figure 48. Link Utilization with upgrade graph


Judging from the above graphs, it is clear that there is an improvement in all application response times, compared to the initial ones, but also in the utilization of the main link, something that makes the expansion of the network's system doable.


In this coursework the use of OPNET was demonstrated in simulating the performance of the hospital's department into three discrete scenarios. The default scenario, the worst case scenario and the expansion scenario. In each one, the topologies and simulation settings were described and the applications response times, as well as the link's utilization, were taken and examined. The need of upgrading the system was emphasized in the case where heavy traffic load was negatively affecting response times, but also in case the network will expand sometime in the future. The practise of upgrading the bandwidth of the links, as well as the CPU's of the servers, proved to be succesfull, as it immediately improved response times in a great amount. However, the exact cost of these changes is not covered on this report and remains for further investigation.


[1] OPNET Modeler software.

[2] Prokkola, J. (2006). OPNET - Network Simulator. VTT Technical Research Centre of Finland, University of Oulu