Dynamic Migration of SDN-based Switches to Distribute Control Layer Load and Increase Efficiency Using Ryu Controller

Subject Areas : Multimedia Processing, Communications Systems, Intelligent Systems

neda abbasi ¹ (Mcs Student, Computer Engineering, Faculty of Electrical and Computer Engineering, Zanjan Branch, Islamic Azad University, Zanjan, Iran.)
Shiva Karimi ² (Assistant Professor, Faculty of Electrical and Computer Engineering, Zanjan Branch, Islamic Azad University, Zanjan, Iran.)

Received: 2022-10-04 Accepted : 2023-03-10 Published : 2022-12-22

Keywords: control layer, software-based, distribution and load balancing, Ryu controller, Dynamic migration of switches,

Abstract :

Introduction: Network needs in today's era include comprehensive improvement of communication and connections with more bandwidth, minimum delay and high throughput. Software-defined networking is one of the most promising modes of future Internet development due to their centralized planning and management capabilities. The load balancing problem for distributed SDN controllers is one of the fundamental challenges of these networks, and a single centralized controller may cause reliability and scalability problems, and although several controllers can solve these problems, a flexibility mechanism is required for load balancing.Method:Therefore, in this research, a load balancing mechanism is proposed based on the load information strategy, so that each controller in the system periodically reports its load information to other controllers, and the overload controllers of the plan, load infomation They do not collect all other controllers before local decisions are made, which in turn reduces controller resource consumption. A load balancing strategy is proposed in the proposed response time-based scheme for multiple SDN controllers, which takes into account the real-time response time characteristics that change with the controller load. Another key part of the proposed scheme is that all these things are done in the Ryu controller by choosing an appropriate response time threshold and processing multiple overload controllers simultaneously, which can solve the load balancing problem of the SDN control panel of the overload controller well. slow This model proposes to transfer the management of specific switches between controllers.Results: In the simulation, the historical load and the future load of the switches were considered, and a switch migration algorithm with double weight was proposed, which led to a decrease in the frequency of switch migration. Experiments have proven that this scheme can quickly provide load balancing between controllers and reduce the number of switch migrations. The results showed that this design can lead to the migration of the overload controller and quickly reduce the workload of the overload controller. As a result, the proposed scheme can effectively achieve load balancing of multiple SDN controllers with high speed. 

References:

1.  Rana, D.S., S.A. Dhondiyal, and S.K. Chamoli, Software defined networking (SDN) challenges, issues and solution. International journal of computer sciences and engineering, 2019. 7(1): p. 884-889.


2. Mohammadi, H. and S. Mostafavi, A Prediction-Based Load Distribution Approach for Software-Defined Networks. Nashriyyah -i Muhandisi -i Barq va Muhandisi -i Kampyutar -i Iran, 2022. 4(19): p. 289-301.


3.  Srivastava, V. and R.S. Pandey, Load balancing for software-defined network: a review. International Journal of Computers and Applications, 2021: p. 1-14.


4.  Deng, X., et al. Efficient SDN Controller Load Balancing Using Deep Reinforcement Learning. in 2021 4th International Conference on Hot Information-Centric Networking (HotICN). 2021. IEEE.


5.  Yeo, S., et al., Achieving Balanced Load Distribution with Reinforcement Learning-Based Switch Migration in Distributed SDN Controllers. Electronics, 2021. 10(2): p. 162.


6.  Malbašić, T., et al., Hybrid SDN Networks: A multi-parameter server load balancing scheme. Journal of Network and Systems Management, 2022. 30(2): p. 1-28.


7.  Salehi, M. and M.A. Irandoost, A Learning Automata Approach for Load Balancing in Software-Defined Networks. International Journal of Smart Electrical Engineering, 2022.


8. Liu, Y., et al. CASM: A Cost-Aware Switch Migration Strategy for Elastic Optical Inter-Datacenter Networks. in Photonics. 2022. Multidisciplinary Digital Publishing Institute.


9.  Li, Z., et al. SDN controller load balancing based on reinforcement learning. in 2018 IEEE 9th International Conference on Software Engineering and Service Science (ICSESS). 2018. IEEE.


10.   Zhong, H., et al., Prediction-based dual-weight switch migration scheme for SDN load balancing. Computer Networks, 2022: p. 108749.


11.  Cabarkapa, D. and Rancic, Performance Analysis of Ryu-POX Controller in Different Tree-Based SDN Topologies. Advances in Electrical and Computer Engineering, 2021. 21(3): p. 31-38.


12. J. Cui, Q.L., H. Zhong, M. Tian, L. Liu, A load-balancing mechanism for distributed SDN control plane using response time. IEEE Trans. Netw. Serv.Manag. 15 (4) (2018) 1197–1206., 2018.