Measuring Internet Bottlenecks: Location, Capacity, and Available Bandwidth
The ability to measure the location, capacity and available bandwidth of bottleneck in end-to-end network path is of major importance in congestion control, streaming applications, quality-of-service, overlay network and traffic engineering. Existing algorithms either fail to measure all the three bottleneck properties, or generate a large amount of probing packets. In addition, they often require deployment in both end hosts. A novel technique, called BNeck, is presented in this paper. It allows end users to efficiently and accurately measure the three bottleneck properties. The key idea of BNeck is that the per-link dispersion of probing packet train can be applied to measure the properties of congested links. The accuracy and efficiency of BNeck have been verified with elaborately designed simulation. The simulation result indicates that various applications can adopt BNeck to probe for the three bottleneck properties without loss of performance.
Unable to display preview. Download preview PDF.
- 1.Carter, R.L., Crovella, M.E.: Measuring bottleneck link speed in packet-switched networks. Perform. Eval. 27(28), 297–318 (1996)Google Scholar
- 3.Hu, N., Li, L.: Locating Internet Bottlenecks: Algorithms, Measurements, and Implications. In: ACM SIGCOMM, August 2004, pp. 41–54 (2004)Google Scholar
- 4.Jacobson, V.: Pathchar - a tool to infer characteristics of Internet paths. Presented in April 1997 MSRI talk (1997)Google Scholar
- 5.Dovrolis, C., Ramanathan, P., Moore, D.: What do packet dispersion techniques measure? In: Proc. IEEE INFOCOM, April 2001, pp. 905–914 (2001)Google Scholar
- 7.RFC 792. Internet control message protocol (September 1981) Google Scholar
- 8.Kapoor, R.: CapProbe: A Simple and Accuracy Capacity Estimation Technique. In: ACM SIGCOMM, August 2004, pp. 67–78 (2004)Google Scholar
- 9.Zhang, Y., Duffield, N.: On the constancy of Internet path properties. In: Proc. ACM SIGCOMM Internet Measurement Workshop, November 2001, pp. 197–211 (2001)Google Scholar
- 11.Taqqu, M.S., Willinger, W.: Proof of a Fundamental Result in Self-Similar Traffic Modeling. ACM Computer Communications Review, 5–23 (April 1997)Google Scholar