An engineering approach to dynamic prediction of network performance from application logs

ABSTRACT

Network measurement traces contain information regarding network behavior over the period of observation. Research carried out from different contexts shows predictions of network behavior can be made depending on network past history. Existing works on network performance prediction use a complicated stochastic modeling approach that extrapolates past data to yield a rough estimate of long-term future network performance. However, prediction of network performance in the immediate future is still an unresolved problem. In this paper, we address network performance prediction as an engineering problem. The main contribution of this paper is to predict network performance dynamically for the immediate future. Our proposal also considers the practical implication of prediction. Therefore, instead of following the conventional approach to predict one single value, we predict a range within which network performance may lie. This range is bounded by our two newly proposed indices, namely, Optimistic Network Performance Index (ONPI) and Robust Network Performance Index (RNPI). Experiments carried out using one-year-long traffic traces between several pairs of real-life networks validate the usefulness of our model.

REFERENCES

• 1: D. Chakraborty , A. Ashir , T. Suganuma , G. Mansfield Keeni , T. K. Roy , N. Shiratori, Self-similar and fractal nature of internet traffic, International Journal of Network Management, v.14 n.2, p.119-129, March 2004 [doi>10.1002/nem.512]
• 2: 2. Ahmed A, Mansfield G, Shiratori N, A meeting scheduling system for global events on the Internet. Proceedings of the Internet Society's 8th Annual Networking Conference, INET'98, July 1998, Geneva, http://www.isoc.org/inet98/proceedings/1b/1b_ 1.html.
• 3: 3. Claffy KC. Internet measurement: Myths about internet data, in Proceedings of the 24th North American Network Operators' Group (NANOG24), February 2002.
• 4: 4. IETF-IPPM-WG: Framework for IP performance metrics, http://www.ietf.org/html.charters/ippm-charter.html.
• 5: 5. Paxson V, Almes G, Mahdavi J, Mathis M. Framework for IP performance metrics. ftp://ftp.isi.edu/ innotes/rfc2330.txt May 1998.
• 6: 6. Koide K, Keeni GM, Kitagata G, Shiratori N. DCAA: a dynamic constrained adaptive aggregation method for effective network traffic information summarization. IEICE Transactions on Communications 2004;E87-B:413-420.
• 7: 7. Paxson V, Almes G, Mahdavi J, Mathis M. An architecture for large-scale Internet measurement. IEEE Communications 1998;36(8):48-54.
• 8: 8. Graham ID. Non-intrusive and accurate measurement of unidirectional delay and delay variation on the Internet. Proceedings of the Internet Society's 8th Annual Networking Conference, INET'98, July 1998, Geneva, http://www.isoc.org/inet98/proceedings/ 6g/6g_2.html.
• 9: 9. Murayama Y, Yamaguchi S. DBS: a powerful tool for TCP performance evaluations. Proceedings of SPIE Volume 3231, November 1997.
• 10: 10. Ashir A, Mansfield G, Shiratori N. Estimation of network characteristics and its use in improving performance of network applications. IEICE Transactions 1999;E82-D:747-755.
• 11: 11. Ashir A, Mansfield G, Shiratori N. Network processes: Scheduling and server selection for efficient operation. Proceedings of Internet Conference 98: IC98. December 1998, Kyoto, Japan.
• 12: 12. Raisanen V, Grotefeld G, Morton A. Network performance measurement with periodic streams. RFC 3432, 2002.
• 13: 13. Squid Internet Object Cache, http://www.nlanr.net.
• 14: 14. Kamiya H, Ohta K, Kato N, Mansfield G, Nemoto Y. Improving efficiency of network services. Asia-Pacific Network Operations and Management Symposium, Sep. 1998.
• 15: 15. PTOPOMIB Working Group (concluded), Physical Topology MIB, http://www.ietf.org/html.charters/ ptopomib-charter.html.
• 16: S. Seshan , M. Stemm , R. H. Katz, SPAND: Shared Passive Network Performance Discovery, University of California at Berkeley, Berkeley, CA, 1997
• 17: 17. Mansfield G, Jayanthi K, Ashir A, Shiratori N. Network maps: Synthesis and applications. International Conference, APSITT99, August 1999, Mongolia.
• 18: 18. Groschwitz NK, Polyzos GC. Atime series model of long-term NSFNET backbone traffic. Proceedings of IEEE ICC'94.
• 19: 19. Baryshnikov Y, Coffman E, Rubenstein D, Yimwadsana B. Traffic prediction on the Internet. Technical Report EE200514-1, Computer Networking Research Center, Columbia University, May 2002.
• 20: 20. Drossu R, Lakshman TV, Obradovi'c Z, Raghavendra C. Single and multiple frame video traffic prediction using neural network models. Computer Networks, Architecture and Applications, Raghavan SV, Jain BN (eds), Chapman and Hall, 1995, pp. 146-158.
• 21: Salahuddin Muhammad Salim Zabir , Ahmed Ashir , Norio Shiratori, Estimation of Network Performance: An approach based on Network Experience, Proceedings of the The 15th International Conference on Information Networking, p.657, January 31-February 02, 2001
• 22: 22. Jain R. The Art of Computer Systems Performance Analysis, Wiley: U.S.A., 1991.
• 23: 23. Ashir A, Suganuma T, Kinoshita T, Roy TK, Mansfield G, Shiratori N. Network traffic characterization and network information services-R and D on JGN. Computer Communications 2001;24: 1734-1743.

INDEX TERMS

Primary Classification:
  C. Computer Systems Organization
  C.2 COMPUTER-COMMUNICATION NETWORKS
      C.2.3 Network Operations
          Subjects: Network management

Additional Classification:
  C. Computer Systems Organization
  C.2 COMPUTER-COMMUNICATION NETWORKS
      C.2.3 Network Operations
          Subjects: Network monitoring