Which Switch to Collect Statistics in Software-Defined Networks

Jun Zhao,Congxiao Bao,Xing LI
DOI: https://doi.org/10.1007/s40009-020-00974-4
2020-01-01
National Academy Science Letters
Abstract:Many monitoring methods have been proposed, due to that network measurement is essential in Software-Defined Networking. However, periodically or adaptively collecting statistics from software switches using per-flow queries incurs significant communication cost and thus increases the loads of switches. In this letter, we propose OpenCost, an approach to decide which switch to collect statistics in Software-Defined Networks based on nonlinear integer programming (NLIP). Simulation results show that OpenCost can reduce the communication cost by 55% in average. OpenCost is an approach based on nonlinear integer programming algorithm, which can be used to reduce communication cost induced by network measurement in Software-Defined Networking. Experiments show that OpenCost can be beneficial to save operating cost for Internet service provider.
What problem does this paper attempt to address?