A Scalable and Efficient Port-Based Adaptive Resource Monitoring Approach in Software Defined Networks

Abstract

The need for communication tools is increasing from the past to the present, therefore the tools and methods in network technologies are evolving. Depending on this development, network scale and complexity increase and the limitations of traditional network technology are surfaced. Software-defined Network (SDN) provides a variety of opportunities for the management and optimization of these challenges. Network resource monitoring is very important for providing information to network applications. This study introduces an adaptive port-level bandwidth monitoring method designed for SDN networks. The proposed approach uses a polling-based monitoring paradigm. There is a trade-off between measurement accuracy and monitoring overhead. With this adaptive method, it is aimed to decrease the overhead while maintaining an acceptable level of accuracy of the measurements and also to use network resources more efficiently. The proposed adaptive monitoring approach has 46% less overhead than the periodic polling method and 6.7% less overhead than the PayLess approach. At the same time, this approach is 5.4% more accurate than the periodic polling approach.