Estimating SDN traffic matrix based on online adaptive information gain maximization method

Abstract

Traffic Matrix (TM) estimation is important for network management and traffic engineering. However, current estimation methods are insufficient in estimation accuracy and measurement cost. In this paper, by using the flow measurement capability in Software Defined Networks (SDN), we propose an Online Information Gain Maximization based SDN traffic matrix estimation method IGME. IGME uses the information gain metric to determine which flows are most informative, and then constructs the measurement flow set iteratively until the accuracy requirement is satisfied or the measurement resource constraint is reached. The experiment results on three Internet measurement datasets show that IGME can improve the estimation accuracy only by consuming a small amount of measurement resource. Besides, the iteration feature of IGME provides a means to dynamically adjust the measurement flow selection choice, so as to adapt to the time-varying characteristics of the network traffic.