| Titre : | Adaptive Load Balancing Based on Artificial Neural Network For Software Defined Networks |
| Auteurs : | Wail Aymen Farhi, Auteur ; Soheyb Ayad, Directeur de thèse |
| Type de document : | Monographie imprimée |
| Editeur : | Biskra [Algérie] : Faculté des Sciences Exactes et des Sciences de la Nature et de la Vie, Université Mohamed Khider, 2019 |
| Format : | 1 vol. (72 p.) / ill. / 29 cm |
| Langues: | Anglais |
| Résumé : |
Traditional networking architectures have many significant limitations that must be overcome to meet modern IT requirements. To overcome these limitations; Software Wined Networking (SDN) is taking place as the new networking approach. One of the major issues of traditional networks is that they use static switches that cause poor utilization of the network resources. Another issue is the packet loss and delay case of switch breakdown. This work proposes a creation of a Adaptive Load aalancing Based on Artificial Neural Network model on SDN network to overcome these issues. A. test-bed has been implemented using Mininet software to emulate he network, and Ryu platform as SDN controller. Python programming language used to define a fat-tree network topology and to write the Controller program, Also to create an Artificial Neural Network model Finally, The result of predictive best path has being showed according the state of the network |
| Sommaire : |
SDN In General 12 1.1 Introduction 12 What is Computer Networking? 12 1.2.1 Traditional N(twork architecture 13 1.2.2 Traditional Network limits 13 1.3 History and Evolution of Software Defined Networking 15 1.3.1 Central Network Control 15 1.3.2 Network Programmability 16 1.3.3 Network Virtualization 16 1.4 Software Defined Networking 17 1.4.1 Introduction to SoftwareDefined Networking 17 1.4.2 SDN -Architecture 17 1.4.3 OpenFlow 21 1.4.4 Migration from Legacy to SDN Networks 26 1.4.5 Conclusion 26 Load Balancing and ANN 27 2.1 Introduction 27 1.9 Load Balancing 27 2.2.1 Round Robin forwarding 27 2.2.2 Random 28 2.2.3 Weighted Round Robin 28 2.2.4 Routing traffic as per the metrics calculated from the traffic. 28 2.3 Network Topology in SDN 29 2.3.1 Fat-Tree topology 29 2.4 Path Computation in SDN 30 2.5 Load Balance in SDN 30 2.6 Artificial Neural Network 31 2.6.1 Applications of Artificial Neural Networks 31 2.6.2 Types of Artificial Neural Network 32 2.6.3 Activation Functions 34 2.7 Network Traffic Measurement 35 2.8 Related Work 36 2.9 Conclusion 37 3 Adaptive Load Balancing Based on ANN 3.1 Introduction 3.2 Work Description 3.2.1 Create Network Topology 3.2.2 Find The Possible Paths 3.2.3 Metrics Collection 3.2.4 DFNN Model Training 3.3 Conclusion 38 4 Implementations and Results 48 4.1 Introduction 48 4.2 Components and Software Tools 48 4.2.1 Mininet 48 4.2.2 Open vSwitch 53 4.2.3 Controller 55 4.2.4 Iperf 58 4.2.5 sFlow 59 4.2.6 Neural network implementation 61 4.2.7 Pre-processing 62 4.2.8 Technical details 62 4.3 Results 63 4.3.1 Conclusion 68 5 General Conclusion 69 5.1 General Conclusion and Future Work 69 |
| Type de document : | Mémoire master |
Disponibilité (1)
| Cote | Support | Localisation | Statut |
|---|---|---|---|
| MINF/449 | Mémoire master | bibliothèque sciences exactes | Consultable |
