A distributed approach to measure IP traffic matrices

The traffic matrix of a telecommunications network is an essential input for any kind of network design and capacity planning decision. In this paper we address a debate surrounding traffic matrix estimation, namely whether or not the costs of direct measurement are too prohibitive to be practical. We examine the feasibility of direct measurement by outlining the computation, communication and storage overheads, for traffic matrices defined at different granularity levels. We illustrate that today's technology, that necessitates a centralized solution, does indeed incur prohibitive costs. We explain what steps are necessary to move towards fully distributed solutions, that would drastically reduce many overheads. However, we illustrate that the basic distributed solution, in which flow monitors are on all the time, is excessive and unnecessary. By discovering and taking advantage of a key stability property underlying traffic matrices, we are able to propose a new scheme that is distributed and relies only on a limited use of flow measurement data. Our approach is simple, accurate and scalable. Furthermore, it significantly reduces the overheads above and beyond the basic distributed solution. Our results imply that direct measurement of traffic matrices should become feasible in the near future.

[1]  Anja Feldmann,et al.  Deriving traffic demands for operational IP networks: methodology and experience , 2000, SIGCOMM.

[2]  B. Yu,et al.  Time-varying network tomography: router link data , 2000, 2000 IEEE International Symposium on Information Theory (Cat. No.00CH37060).

[3]  Carsten Lund,et al.  Properties and prediction of flow statistics from sampled packet streams , 2002, IMW '02.

[4]  Yin Zhang,et al.  BGP routing stability of popular destinations , 2002, IMW '02.

[5]  Kavé Salamatian,et al.  Traffic matrix estimation: existing techniques and new directions , 2002, SIGCOMM '02.

[6]  Christophe Diot,et al.  Taxonomy of IP traffic matrices , 2002, SPIE ITCom.

[7]  Albert G. Greenberg,et al.  Fast accurate computation of large-scale IP traffic matrices from link loads , 2003, SIGMETRICS '03.

[8]  Konstantina Papagiannaki,et al.  Long-term forecasting of Internet backbone traffic: observations and initial models , 2003, IEEE INFOCOM 2003. Twenty-second Annual Joint Conference of the IEEE Computer and Communications Societies (IEEE Cat. No.03CH37428).

[9]  Bin Yu,et al.  Pseudo likelihood estimation in network tomography , 2003, IEEE INFOCOM 2003. Twenty-second Annual Joint Conference of the IEEE Computer and Communications Societies (IEEE Cat. No.03CH37428).

[10]  Kihong Park,et al.  Scalability and traffic control in IP networks , 2001, Computer Communications.

[11]  Carsten Lund,et al.  An information-theoretic approach to traffic matrix estimation , 2003, SIGCOMM '03.

[12]  Emilio Leonardi,et al.  How to identify and estimate the largest traffic matrix elements in a dynamic environment , 2004, SIGMETRICS '04/Performance '04.

[13]  George Varghese,et al.  The measurement manifesto , 2004, Comput. Commun. Rev..