Abstract
The relative performance of different data collection methods in the assessment of various traffic parameters is significant when the amount of data generated by a complete trace of a traffic interval is computationally overwhelming, and even capturing summary statistics for all traffic is impractical. This paper presents a study of the performance of various methods of sampling in answering questions related to wide area network traffic characterization. Using a packet trace from a network environment that aggregates traffic from a large number of sources, we simulate various sampling approaches, including time-driven and event-driven methods, with both random and deterministic selection patterns, at a variety of granularities. Using several metrics which indicate the similarity between two distributions, we then compare the sampled traces to the parent population. Our results revealed that the time-triggered techniques did not perform as well as the packet-triggered ones. Furthermore, the performance differences within each class (packet-based or time-based techniques) are small.
- 1 T.W. Anderson and D.A. Darling. Asymptotic theory of certain goodness-of-fit criteria based on stochastic processes. Annals of Mathematical Statistics, 23:193- 212, 1954.]]Google ScholarCross Ref
- 2 ANS. ARTS: ANSnet Router Statistics software, 1992.]]Google Scholar
- 3 B..T. Braden and A. DeSchon. NNStat: Internet statistics collection package. Introduction and User Guide. Technical Report KR-88-206, ISI, USC, 1988. Available for a-tip from isi.edu.]]Google Scholar
- 4 J.D. Case, M. Fedor, M.L. Schoffstall, and C. Davin. Simple Network Management Protocol (SNMP). Internet Request for Comments Series I~FC 1157, 1987.]] Google ScholarDigital Library
- 5 K. Claffy, H.-W. Braun, and G. C. Polysos. Trig long-term growth of the nsfnet backbone. In P~oc. INET '93, San Francisco, CA, August 1993.]]Google Scholar
- 6 K. Claffy, G. C. Polyzos, and H.-W. Braun. Traffic characteristics of the T1 NSFNET backbone. In Proc. INFOCOM '93, San Francisco, CA, April 1993.]]Google ScholarCross Ref
- 7 W. Cochran. Sampling Techniques. John Wiley Sons, 1987.]]Google Scholar
- 8 It. B. D'Agostino and M. A. Stevens, editors. Goodness of Fit. Marcel Dekker, Inc., 1986.]]Google Scholar
- 9 J. Fleiss. Statistical Methods for Rates and Proportions. John Wiley & Sons, 1981.]]Google Scholar
- 10 L. Goodman and W. Kruskal. Measures of association for cross classifications. Journal of the American Statistical Association, pages 732-763, December 1954.]]Google Scholar
- 11 P.R. Krishnaiah and C.it. Itao. Handbook of Statistics, Volume 6: Sampling. North-Holland, 1988.]]Google Scholar
- 12 V. Paxson. Empirically-Derived Analytic Models of Wide Area TCP Connections. Master's thesis, UC, Berkeley and Lawrence Livermore National Laboratory, 1992.]]Google Scholar
Index Terms
- Application of sampling methodologies to network traffic characterization
Recommendations
Application of sampling methodologies to network traffic characterization
SIGCOMM '93: Conference proceedings on Communications architectures, protocols and applicationsThe relative performance of different data collection methods in the assessment of various traffic parameters is significant when the amount of data generated by a complete trace of a traffic interval is computationally overwhelming, and even capturing ...
Robust network dimensioning for realtime services over IP networks with traffic deviation
Network planning for realtime services in traditional systems is normally performed based on a fixed traffic demand. Realtime services such as voice and video are currently being widely deployed over packet-based IP networks. Due to the lack of resource ...
Comments