Software-Defined Networking 2

Fetia Bannour is an Associate Professor in Computer Science at ENSIIE and SAMOVAR, France. She earned her PhD in Computer Networks from the University of Paris-Est. Her research interests include the control/management of SDN and NFV-based future networks.Sami Souihi is an Associate Professor in Computer Science at the N&T Department and the LiSSi-TincNET research team (Paris-Est Créteil University (UPEC)), France. His research work focuses on adaptive mechanisms in large-scale dynamic systems, among others.Abdelhamid Mellouk is currently the Director of IT4H High School Engineering Department, a Full Professor at the University of Paris-Est Créteil (UPEC) and Head of UPEC-LiSSi-TincNET research team, France. He is the founder of the Network Control Research and Curricula activities in UPEC, the current co President of the French Deep Tech Data Science and Artificial Intelligence Systematic Hub, member of the High Scientific Research and Technology National Council (DZCNRST) and associate editor in several top ranking scientific journals.

• Acronyms ixPreface xiiiIntroduction xviiChapter 1 Toward a Decentralized SDN Control Architecture: Overview and Taxonomy 11.1. Introduction 11.2. Software-defined networking: a centralized control architecture 21.2.1. Conventional networking and the SDN paradigm 21.2.2. The SDN architecture 31.3. Physical classification of existing SDN control plane architectures 81.3.1. Physically centralized SDN control 81.3.2. Physically distributed SDN control 111.4. Logical classification of existing SDN control plane architectures 161.4.1. Logically centralized SDN control 171.4.2. Logically distributed SDN control 221.5. Conclusion 26Chapter 2. Decentralized SDN Control: Major Open Challenges  272.1. Introduction 272.2. Scalability 282.2.1. Data plane extensions 302.2.2. Control plane distribution 322.3. Reliability 332.3.1. Control state redundancy 332.3.2. Controller failover 342.4. Controller state consistency 352.4.1. Static consistency 362.4.2. Adaptive multi-level consistency 372.5. Interoperability 382.5.1. Interoperability between the SDN controllers 382.5.2. SDN interoperability with legacy networks 382.6. Other challenges 392.7. Conclusion 40Chapter 3 Scalability and Reliability Aware SDN Controller Placement Strategies 413.1. Introduction 413.2. Related work 423.3. The SDN controller placement optimization problem 443.3.1. Problem statement 443.3.2. Problem formulation 453.3.3. Placement metrics 453.4. The proposed SDN controller placement scheme 493.4.1. The adopted approach 493.4.2. Multi-criteria placement algorithms 503.4.3. Gradual strategies 523.5. Performance evaluation 533.5.1. Simulation settings 533.5.2. Simulation results 543.6. Discussion 603.7. Conclusion 62Chapter 4 Adaptive and Continuous Consistency for Distributed SDN Controllers: Anti-Entropy Reconciliation Mechanism 654.1. Introduction 654.2. Related work 664.3. The consistency problem in SDN 684.3.1. Consistency trade-offs in SDN 684.3.2. Consistency models in SDN 694.4. Consistency models in ONOS 704.4.1. Strong consistency in ONOS 704.4.2. Eventual consistency in ONOS 714.5. The proposed adaptive consistency for ONOS 724.5.1. A continuous consistency model for ONOS 724.5.2. Our consistency adaptation strategy for ONOS 744.5.3. Our implementation approach 744.6. Performance evaluation 764.6.1. Experimental setup 764.6.2. Results 764.7. Conclusion 79Chapter 5 Adaptive and Continuous Consistency for Distributed SDN Controllers: Quorum-Based Replication 815.1. Introduction 815.2. Background on eventual consistency in distributed data stores 835.2.1. Consistency and performance metrics 835.2.2. Adaptive consistency control 845.2.3. Existing modern tunable consistency systems 845.3. The proposed adaptive Quorum-inspired consistency for ONOS 865.3.1. A continuous consistency model for ONOS 865.3.2. Our Quorum-inspired consistency adaptation strategy for ONOS 875.4. Implementation approach on ONOS 935.4.1. Design of a CDN-like application 935.4.2. State synchronization and content distribution 945.4.3. Content delivery to customers 955.5. Performance evaluation 975.5.1. Application-specific performance and consistency metrics 975.5.2. Experimental setup 985.5.3. Results 1035.6. Conclusion 112Conclusions and Perspectives 115C.1. Summary of contributions 115C.2. Perspectives and future work 117References 121Index 139