OpenFlow-based Low-Overhead and High-Accuracy SDN Measurement Framework.

Qiang He,Xingwei Wang,Min Huang
DOI: https://doi.org/10.1002/ett.3263
IF: 3.6
2018-01-01
Transactions on Emerging Telecommunications Technologies
Abstract:Network measurement is an essential part of network management. Software-defined networking (SDN) is a promising networking architecture to make network management more flexible by separating the control plane and the data plane. Recently, the corresponding research studies about SDN measurement have not attracted much attention due to its preliminary stage, especially the currently proposed SDN measurement methods that have been suffering from high network overhead, low accuracy, and incompatibility. To address these issues, this paper proposes an OpenFlow-based low-overhead and high-accuracy (OpenLh) SDN measurement framework. Firstly, to decrease the load of edge switch and network overhead, we propose a switch querying algorithm to measure the first and last switches per flow periodically. Secondly, a novel adaptive polling algorithm is designed to adjust the polling interval, which reduces network overhead while optimizing measurement accuracy. Furthermore, we quantitatively analyze the trade-off among measurement overhead, accuracy, and minimum time interval. Finally, the POX controller is utilized to implement OpenLh, and its effectiveness is verified by Mininet. The simulation results demonstrate that OpenLh can obtain lower network overhead without compromising measurement accuracy than the chosen benchmarks.
What problem does this paper attempt to address?