Abstract | ||
---|---|---|
Distributed architectures of the SDN control plane require a careful design for balancing the load among controllers. Solutions proposed for SDN load balancing usually use switch migration operations. However, an efficient switch migration means triggering the operation at the right moment, and judiciously choosing the migrated switch and the destination controller. Here, we propose a switch migration scheduling algorithm to improve the migration efficiency, and ensure load balancing between controllers. Our algorithm uses a multi-step ARIMA forecasting model to predict the long-term controllers load. When an overload is predicted, a switch migration operation is scheduled in advance. After validating the accuracy of the ARIMA forecasting model, we evaluated the performance of the algorithm by analyzing the response time of controllers. Numerical results confirm the performance of the proposed algorithm. |
Year | DOI | Venue |
---|---|---|
2019 | 10.1109/ICC.2019.8761469 | ICC 2019 - 2019 IEEE International Conference on Communications (ICC) |
Keywords | Field | DocType |
ARIMA forecasting model,SDN load balancing,SDN control plane,destination controller,prediction-based switch migration scheduling algorithm | Routing control plane,Control theory,Load balancing (computing),Scheduling (computing),Computer science,Response time,Real-time computing,Autoregressive integrated moving average | Conference |
ISSN | ISBN | Citations |
1550-3607 | 978-1-5386-8089-6 | 0 |
PageRank | References | Authors |
0.34 | 9 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Abderrahime Filali | 1 | 1 | 1.36 |
Soumaya Cherkaoui | 2 | 187 | 40.89 |
Abdellatif Kobbane | 3 | 126 | 27.54 |