ABSTRACT
Modern campus networks are designed for high-speed service with the capacity to handle many organization’s bandwidth requirements, support data applications such as electronic-mail, web and files sharing services. These networks sometime lack the capacity to guarantee quality of service for voice, video and other interactive applications due to bandwidth requirement and high packet end-to-end delay. In some networks, voice, video and other interactive packets must be given priority in terms of bandwidth and delay over less-time-sensitive packets such as mail or file sharing for the networks to meet the QoS demand of these voices, video and other interactive applications. In this research, best-effort service model currently employed in Ahmadu Bello University (ABU) Internet network and differentiated services networking model define in RFC 2474 were analyzed. Enhanced differentiated services model was developed to improve and achieve better reduction in packet end-to-end delay of interactive traffic such as video conferencing and voice streaming in networks. The simulation results obtained of simulation of the three services model showed that enhanced differentiated services model with mean packet end-to-end of 0.10481s has 26% delay reduction as compared with differentiated services model with mean packet end-to-end of 0.10757s. The enhanced differentiated services model has 32% delay reduction as compared to the best-effort services model with mean packet end-to-end of 0.15417s. Similarly, validation results obtained of mean delay of the enhanced differentiated services model performed better than the differentiated services model by and better than the best-effort services model by 83%. These results demonstrate the superiority of the enhanced differentiated service to differentiated service and best-effort services models in reduction of packet end-to-end delay of a network.
TABLE OF CONTENTS
Title Page . . . . . . . . . . i Declaration . . . . . . . . . . ii Certification . . . . . . . . . . iii Dedication . . . . . . . . . . iv Acknowledgement . . . . . . . . . v Table of Contents . . . . . . . . . vi List of Tables . . . . . . . . . ix List of Figures . . . . . . . . . x List of Abbreviations . . . . . . . . . xi List of Appendices . . . . . . . . . xii Abstract . . . . . . . . . . xiii CHAPTER ONE: INTRODUCTION 1.0 Background . . . . . . . . . 1
1.1.2 Ahmadu Bello University Internet Network . . . . 2
1.2 Motivation . . . . . . . . . 4
1.3 Significance of Research . . . . . . . 5
1.4 Statement of Problem . . . . . . . 5
1.5 Aim and Objectives . . . . . . . . 6
1.6 Methodology . . . . . . . . . 6
1.7 Dissertation Organisation . . . . . . . 7
CHAPTER TWO: LITERATURE REVIEW
2.1 Introduction . . . . . . . . . 8
2.2 Review of Fundamental Concepts . . . . . . 8
2.2.1 Campus Area Network . . . . . . . 8
2.2.1.1 Features Campus Network . . . . . . 10
2.2.1.2 Campus Network Architecture . . . . . . 11
2.2.2 Performance Metrics in Computer Communication Networks . . 13
7
2.2.3 Need for Quality of Service . . . . . . 14
2.2.3.1 Factors which Determine Quality of Service . . . . 15
2.2.4 Network Services Models . . . . . . 16
2.2.4.1 Best-Effort Services Model . . . . . . 16
2.2.4.2 Integrated Services Model . . . . . . 17
2.2.4.3 Differentiated Services Model . . . . . . 18
2.2.4.4 Implementation of Differentiated Services . . . . 20
2.2.4.5 Analysis of Assured Forwarding . . . . 26
2.2.4.6 Limitations of Differentiated Services Architecture . . . 28
2.3 Review of Similar Research Works . . . . . . 29
2.3.1 Summary of Literature Review . . . . . . 34
CHAPTER THREE: MATERIALS AND METHODS
3.1 Introduction . . . . . . . . . 35
3.2 Implementation of Differentiated Services Model Solution . . . 35
3.2.1 Implementation of Enhance Differentiated Services . . . 36
3.3 Simulation of Best-Effort and Developed Differentiated Services Model . 37
3.3.1 Best-Effort Network Service Model Simulation . . . . 37
3.3.2 Differentiated Services Models Simulation . . . . 39
3.3.3 Enhanced Differentiated Services Network Models Simulation . 43
3.4 Application and Validation of Enhanced Differentiated services model . 46
3.4.1 Best-Effort Service Model Validation . . . . . 47
3.4.2 Differentiated Service Model Validation . . . . . 48
3.4.3 Enhanced Differentiated Service Model Validation . . . 49
8
CHAPTER FOUR: RESULTS AND DISCUSSIONS
4.1 Introduction . . . . . . . . . 51
4.2 Analysis of Results and Data Obtained using OPNET Simulator . . 51
4.2.1 Summary of Results Obtained using OPNET Simulator . . . 56
4.3 Results of Validation and Application of Enhanced Differentiated Model . 56
CHAPTER FIVE: SUMMARY, CONCLUSION AND RECOMMENDATION
5.1 Summary of findings . . . . . . . . 59
5.2 Conclusion . . . . . . . . . 59
5.3 Limitation . . . . . . . . . 60
5.4 Suggestion for Further Work . . . . . . . 60
REFERENCES . . . . . . . . . 61
CHAPTER ONE
INTRODUCTION
1.1 Background
Information technology is strategically important to the goals and aspirations of business enterprises, government entities and educational institutions, particularly Universities. It is the cornerstone that enables the University’s faculties, researchers, students, administrators, and staff to discover, learn, reach out, and serve humanity. Campus Area Networks (CANs) such as ABU network transmit interactive and multimedia applications such as video and audio streaming for experimental, practical and other uses. The numbers of such interactive and multimedia applications on such networks are on the increase. Voice, video and data applications demand different types of performance assurance and so Quality of Service (QoS) provision is one of the important components in the design of such networks. Researchers have done considerable work in developing QoS models, mechanism and queuing disciplines to improve transmission of interactive and multimedia applications on IP networks. The challenge of implementation of the developed QoS models, and queuing mechanism on last mile networks of the internet and the inability and inappropriate implementation of most of the QoS models result in the degradation of the QoS in terms of packet loss, packet delays, and packet delay variation for messages transmitted over the networks.
Packet-based networks are networks in which packets sent from a source may traverse different paths to arrive at the final destination. The packets that are routed over separate paths are reassembled at the destination. Transmission rates of the various paths may vary depending upon the usage of the network paths over which the packets are being transmitted (Abaye et al., 2006).
15
During heavy traffic conditions, packets may be delayed and lost. Packet delays and losses cause poor performance of the network and are more obvious with voice and other interactive streaming communications. Interactive streaming packets (voice, multimedia) in a network share the network bandwidth with conventional non-streaming packets (such as data associated with electronic mail, file transfer, web access, and other traffic). Voice data and other interactive packets that are lost or delayed due to inadequate or unavailable capacity of networks may result in gaps, silence, and clipping of audio at the receiving end, thus affecting the QoS of the network (Abaye et al., 2006). One solution for this problem is the over provisioning of bandwidth. Howeve is an expensive option for service utilises and can be difficult to ensure in all cases. Another and a preferred solution is the use of QoS models. Two of such models that have been standardized by the Internet Engineering Task Force (IETF), are integrated services and differentiated services models. Due to scalability issue with integrated service model, differentiated services models is mostly preferred. The preferred differentiated services model also has the challenge of Packet being processed and eventually dropped during the last process, if threshold of queues are exceeded. This leads to added delay to incoming packet caused by the processing time of packets which are eventually dropped, hence the need to improve the differentiated services models for better performances.
1.7.2 Ahmadu Bello University Campus Network
The physical connectivity diagram of ABU network shown in Figure 1.1 illustrates how network devices are connected. Packets are transmitted between devices through a series of routers. Each intermediate router in the network may receive packets via multiple data streams that are routed simultaneously from their source devices to their respective destinations. In such conditions, packets may have to be stored at the intermediate routers for transmission at a later time due to reasons such as full buffers, packet loss or scheduling.
16
In ABU network, arriving internet packets are assigned on per connection queuing using first in first out (FIFO) queuing disciple.
Figure 1.1: Topology Diagram of ABU Network (Tekanyi 2014)
17
The Per Connection Queuing (PCQ) is a method that can be used to dynamically equalize or shape traffic for multiple users. It is possible to divide PCQ scenarios into three major groups: equal bandwidth for a number of users, certain bandwidth equal distribution between users, and unknown bandwidth equal distribution between users. The ABU network uses PCQ with best-effort services model (without QoS), a staff is allotted 1Mbit/s bandwidth and a student allotted 50kbit/s bandwidth when there is congestion, while during light periods of traffic when there is no congestion at the interfaces, packet on arrival are served and dispatched. During heavy load periods of traffic in which packets arrival rate are faster than outgoings from the interface, congestion occurs and packets queue for services to avoid drop, once the interface is free they are serviced and transmitted based on their assigned priority employed at the interface. On such instances of congestion, transmission of multimedia and interactive traffic may suffer degradation of services across the network since such traffics has high bandwidth requirement and are delays sensitive. To avoid degradation of services of such multimedia and interactive traffics, service differentiation and priority will require being given to the multimedia and interactive traffic over non delay sensitive traffic at the interface with congestion. Therefore there is a need to continuously improve the ABU network by evolving better QoS model to support and improve the transmission and deployment of interactive applications on the network.
1.8 Motivation
Transmission of voice, video and other interactive traffic may suffer degradation of services across a campus Internet network if there is congestion. Voice, video and data applications demand different types of performance assurance and QoS. Campus Internet network may require service differentiation and priority queuing at the interface in which congestion occurs to
18
give priority to delay sensitive traffic such as voice, video and other interactive traffic over non delay sensitive traffics such as file download and electronic mail traffic. The motivation for this dissertation is to develop and improve the differentiated services QoS architecture to reduce packet end to end delay and guarantee QoS for of delay sensitive traffic on campus Internet network.
1.9 Significance of Research
The contributions of this research are as follows:
i. Improve packet end to end delay by 32% compared with ABU current services model.
ii. Achieved 26% packet end to end delay reduction when compared with standard differentiated services model.
1.10 Statement of Problem
Due to the demand for new types of services, Universities’ campuses are facing new challenges related to network infrastructure. The recent push to converge voice and video onto the data network has made QoS guarantee a critical factor in the design and engineering of campus area networks. Transmission of voice, video, and other interactive traffic may suffer degradation of services across the network if there is congestion and may require service differentiation and priority queuing at the interface where congestion occurs in order to give priority to delay sensitive over non delay sensitive traffic. Hence, the need to develop and continuously improve QoS architecture that can differentiate and give priority to delay sensitive video, voice and other interactive traffic during congestion periods in a campus network in order to reduce packet end to end delay of delay sensitive traffic.
19
1.11 Aim and Objectives
The aim of this research is to develop an enhanced differentiated services model to improve the differentiated services model define in RFC 2474. This is with the view to reduce packet delay of sensitive traffic on a campus Internet network. The objectives of this research are:
i. Develop an enhance differentiated services model for a Campus Area Network (CAN)
ii. Compare the proposed enhanced differentiated services with best-effort and differentiated service model using Riverbed Modeller Academic Edition 17.5 (OPNET)
iii. Validate the developed enhanced differentiated services model using live routers on a test-best network to establish the improvement achieved by the developed model.
1.12 Methodology
The research methodology is as follows:
i. Implementation of differentiated services model and its simulation using OPNET simulator.
ii. Development of enhanced differentiated services model and its simulation using OPNET simulator.
iii. Comparison of performance of best-effort, differentiated services model with that of the developed enhanced differentiated services model.
iv. Validation of the developed enhanced differentiated services model using live routers and switches on a developed Test-bed network.
20
1.13 Dissertation Organisation
The summary of how the rest of the chapters are organised is presented thus: Having presented the introduction with the general background of the dissertation in chapter one, chapter two presents the literature review consisting of systematic review of some fundamental concepts theories relevant to differentiated services model and the developed enhanced differentiated service model. Also in this chapter is a comprehensive review of work similar to this dissertation. Development of the enhanced differentiated services model is analysed in chapter three. Prior to this, the general structure of the differentiated services model defined in RFC 2474 and it shortcoming is briefly highlighted in the chapter. Chapter four presents the results of simulation and validation of the developed differentiated services model. A test bed using live router and switches was developed and the best-effort, differentiated and the developed enhanced differentiated service were configured and the end to end delay measured using Wireshark network analyser. Chapter five presented the conclusion of the study. Also in the chapter are the summary of finding and the suggestion for further work. Finally, all cited references are presented at the end of this dissertation work.
21
IF YOU CAN'T FIND YOUR TOPIC, CLICK HERE TO HIRE A WRITER»
