Overview of Monitor Selection in Computer Networks

Abstract

With the development of the computer networks, network measurement becomes more and more important and complicated. It may need to deploy many monitors and send many packets, which will introduce impact on the performance of the network. So, it is very meaningful to design network measurement architecture to get as much as the network information using as little as the measurement cost with sophisticated monitors selection, which is referred as the problem of monitor selection. The problem of monitor selection for flow and failures as well as delay is related in this paper. The technologies are dived into three categories which are based on the Graph Theory and the Mathematical Programming as well as the Group Betweenness Centrality theory. This division is coincident with the characteristics of the monitor selection technologies. Then, we go into the main research methods and results referring to each of the three categories. At last, we talk about the further research direction regarding the shortcomings of the present methods. © Springer-Verlag Berlin Heidelberg 2013.