Abstract
The Internet is a very large scale, complex, dynamical system that is hard to model and analyze. In this paper, we develop and analyze statistical models for the observed end-to-end network performance based on extensive packet-level traces (consisting of approximately 1.5 billion packets) collected from the primary Web site for the Atlanta Summer Olympic Games in 1996. We find that observed mean throughputs for these transfers measured over 60 million complete connections vary widely as a function of end-host location and time of day, confirming that the Internet is characterized by a large degree of heterogeneity. Despite this heterogeneity, we find (using best-fit linear regression techniques) that we can express the throughput for Web transfers to most hosts as a random variable with a log-normal distribution. Then, using observed throughput as the control parameter, we attempt to quantify the spatial (statistical similarity across neighboring hosts) and temporal (persistence over time) stability of network performance. We find that Internet hosts that are close to each other often have almost identically distributed probability distributions of throughput. We also find that throughputs to individual hosts often do not change appreciably for several minutes. Overall, these results indicate that there is promise in protocol mechanisms that cache and share network characteristics both within a single host and amongst nearby hosts.
- 1 K. Agrawala and D Sanghi. Network Dynamics: An Experimental Study of the Interact. In Proc. GLOBECOM "92, December 1992.Google Scholar
- 2 IBM AlphaWorks Home Page. http:// www.alphaworks.ibm.com, 1996.Google Scholar
- 3 M Arlitt and C.L. Williamson. Web Server Workload Characterization: The Search for Invariants. In Proc. ACM SIGMETR1CS '96, May 1996. Google Scholar
Digital Library
- 4 J.C Bolot. End-to-End Packet Delay and Loss Behavior in the Internet. In Proc. ACM SIG- COMM '93, San Francisco, CA, Sept 1993. Google Scholar
Digital Library
- 5 R Caceres. Multiplexing Traffic and the Entrance to Wide Area Networks. PhD thesis, University of California at Berkeley, December 1992. Google Scholar
Digital Library
- 6 CAIRN Home Page. http:/lwww.isi.eduldiv7/ cairn/, 1996.Google Scholar
- 7 R.L. Carter and M. E. Crovella. Dynamic server selection using bandwidth probing in wide-area networks. Technical Report B U-CS-96-007, Computer Science Department, Boston University, March 1996. Google Scholar
Digital Library
- 8 M. Crovella and A. Bestavros. Self-Similarity in World Wide Web Traffic: Evidence and Possible Causes. In Proc. ACM SIGMETRICS, May 1996. Google Scholar
Digital Library
- 9 R.B. D'Agostino and M.A. Stephens. Goodness-of-Fit Techniques. Marcel Dekker, New York, NY, 1986. Google Scholar
Digital Library
- 10 P. Danzig and S. Jamin. tcplib: A Library of TCP Internetwork Traffic Characteristics. Technical Report USC-CS-91-495, University of Southern California, 1991.Google Scholar
- 11 P. Danzig, S. Jamin, R. Caceres, D. Mitzel, and D. Estrin. An Empirical Workload Model for Driving Wide-Area TCP/IP Network Simulations. Internetworking: Research and Experience, 3(1 ): 1-26, March 1992.Google Scholar
- 12 O. Gudmundson, D. Sanghi, and K. Agrawala. Experimental Assessment of End-to-End Behavior on Internet. In Proc. InfoComm '93, March 1993.Google Scholar
- 13 S.A. Heimlich. Traffic Characterization of the NSFnet National Backbone. In Proc. SIGMET- RICS '90, May 1990. Google Scholar
Digital Library
- 14 R. Jain. The Art of Computer Systems Performance Analysis. John Wiley and Sons, 1991.Google Scholar
- 15 S. Keshav. Packet-Pair Flow Control. IEEE/ A CM Transactions on Networking, February 1995.Google Scholar
- 16 W. Leland, M. Taqqu, W. Willinger, and D. Wilson. On the Self-Similar Nature of Ethernet Traffic. In Proc. SIGCOMM '93, September 1993. Google Scholar
Digital Library
- 17 B.A. Mah. An Empirical Model of HTTP Network Traffic. In Proc. lnfoComm '97, April 1997. Google Scholar
Digital Library
- 18 S. McCanne and V. Jacobson. The BSD Packet Filter: A New Architecture for User-Level Packet Caplure. In Proc. Winter '93 USENIX Conference, San Diego, CA, January I993. Google Scholar
Digital Library
- 19 J.C. Mogul. Observing TCP Dynamics in Real Networks. Technical Report 92/2, Digital Western Research Lab, April 1992.Google Scholar
- 20 J.C. Mogul. Network Behavior of a Busy Web Server and its Clients. Technical Report 95/5, Digital Western Research Lab, October 1995.Google Scholar
- 21 ns- LBNI, Network Simulator. http://wwwnrg.ee.lbl.gov/ns/, 1996.Google Scholar
- 22 Official 1996 olympic web site- home page. http://www.atlanta.olympic.org, 1996.Google Scholar
- 23 V Paxson. Empirically-Derived Analytic Models of Wide-Area TCP Connections. IEEE/ACM Transactions on Networking, 2(4):316-336, August 1994. Google Scholar
Digital Library
- 24 V. Paxson. An Analysis of End-to-End Internet Dynamics, Part II, 1996. Ph.D. dissertation in preparation. Google Scholar
Digital Library
- 25 V. Paxson. End-to-End Routing Behavior in the Internet. In Proc. ACM SIGCOMM '96, August 1996. Google Scholar
Digital Library
- 26 V. Paxson and S. Floyd. Wide-Area Traffic: The Failure of Poisson Modeling. IEEE/ACM Transactions on Networking, 3(3):226-244, June 1995. Google Scholar
Digital Library
- 27 M. Schwartz, D. Ewing, and R. HaU. A Measurement of Interact File Transfer Traffic. Technical Report CU-CS-371-92, University of Colorado, January 1992.Google Scholar
- 28 W.R. Stevens. TCP/IP Illustrated, Volume 1. Addison-Wesley, Reading, MA, Nov 1994.Google Scholar
- 29 WOM Boiler Room. http://www.womplex.ibm.com, 1996.Google Scholar
Index Terms
Analyzing stability in wide-area network performance
Recommendations
Analyzing stability in wide-area network performance
SIGMETRICS '97: Proceedings of the 1997 ACM SIGMETRICS international conference on Measurement and modeling of computer systemsThe Internet is a very large scale, complex, dynamical system that is hard to model and analyze. In this paper, we develop and analyze statistical models for the observed end-to-end network performance based on extensive packet-level traces (consisting ...
Enhancing TCP performance in wide-area cellular wireless networks: transport level approaches
Wireless communications systems and networksInternet technology-based architectures and protocols for supporting multimedia traffic over wireless networks are evolving. Since TCP (Transmission Control Protocol)/IP (Internet Protocol) is the standard network protocol stack on the Internet, its use ...
SCTP over high speed wide area networks
ICN'05: Proceedings of the 4th international conference on Networking - Volume Part IThe Stream Control Transmission Protocol (SCTP) is a reliable transport protocol to tackle the limitations of TCP and UDP. SCTP was originally designed to transport PSTN signaling messages over IP networks, but is also capable of serving as a general-...






Comments