• Sorted by Date • Classified by Publication Type • Classified by Research Category •
Jia-Yu Pan, Srinivasan Seshan, and Christos Faloutsos. FastCARS: Fast, Correlation-Aware Sampling for Network Data Mining.
In Proceedings of IEEE GlobeCOM 2002 - Global Internet Symposium, 2002.
Taipei, Taiwan, November 17-21, 2002
[PDF]228.2kB [gzipped postscript]129.9kB [HTML]4.2kB
Technology trends are making it more and more difficult to observe and record the large amount of data generated by high speed links. Traffic sampling techniques provide a simple alternative that reduces the volume of data collected. Unfortunately, existing sampling techniques largely hide any temporal relationship in the recorded data.Our proposed method, ``FastCARS'' naturally captures statistics forpackets that are 1, 2 or more steps away. It has thefollowing properties: (a) it provides accurate measurementsof full trace's statistics, (b) it is simple and can be easilyimplemented, (c) it captures correlations between successive packets,as well as packets that are further apart, and (d) it is scalable andflexible such that it can be easily adjusted to take into account priorknowledge about characteristics of particular traces.We also propose several new tools for network data mining that use theinformation provided by FastCARS.The experimental results on multiple, real-world datasets (233Mb intotal), show that the proposed FastCARS sampling method and these new datamining tools are effective. With these tools, we show that theindependence assumption of packet arrival is not correct, and packettrains may not be the only cause of dependence among arrivals.
@InProceedings{GlobalInternet02FastCARS, author = {Jia-Yu Pan and Srinivasan Seshan and Christos Faloutsos}, title = {FastCARS: Fast, Correlation-Aware Sampling for Network Data Mining}, booktitle = {Proceedings of IEEE GlobeCOM 2002 - Global Internet Symposium}, year = 2002, wwwnote = {Taipei, Taiwan, November 17-21, 2002}, abstract = {Technology trends are making it more and more difficult to observe and record the large amount of data generated by high speed links. Traffic sampling techniques provide a simple alternative that reduces the volume of data collected. Unfortunately, existing sampling techniques largely hide any temporal relationship in the recorded data. Our proposed method, ``FastCARS'' naturally captures statistics for packets that are 1, 2 or more steps away. It has the following properties: (a) it provides accurate measurements of full trace's statistics, (b) it is simple and can be easily implemented, (c) it captures correlations between successive packets, as well as packets that are further apart, and (d) it is scalable and flexible such that it can be easily adjusted to take into account prior knowledge about characteristics of particular traces. We also propose several new tools for network data mining that use the information provided by FastCARS. The experimental results on multiple, real-world datasets (233Mb in total), show that the proposed FastCARS sampling method and these new data mining tools are effective. With these tools, we show that the independence assumption of packet arrival is not correct, and packet trains may not be the only cause of dependence among arrivals.}, bib2html_pubtype = {Refereed Conference}, bib2html_rescat = {Network Data Mining}, }
Generated by bib2html (written by Patrick Riley ) on Wed Sep 01, 2004 13:24:30