Do Applications Perform Its Original Design? A Preliminary Analysis from Internet Big Data

  • Lei Qian
  • Yinlong LiuEmail author
  • Yanfei Zhang
Conference paper
Part of the Communications in Computer and Information Science book series (CCIS, volume 557)


Transmission Control Protocol (TCP) is the most widely used Internet protocol in today’s Internet. More and more applications are operating towards over TCP. Every application layer protocol is designed to perform a specific purpose, for example File Transfer Protocol (FTP) is designed for file transfer. Therefore every application layer protocol has different performance expectations. In this paper, we try to observe that whether characteristics existence in different application layer protocols by illustrating our performance evaluation result of five different application layer protocols (HTTP, HTTPS, FTP, SMTP, and SSH) in individual flow level from over 10 years IP packets trace files collected from two different locations. Our results show that the performance of application layer protocol is still in chaotic and some result is difficult to find a reasonable explanation. We believe that this report is a starting point for both researchers and Internet participators to explore possible reasoning behind of the results.


  1. 1.
    Qian, L., Carpenter, B.E.: A flow-based performance analysis of TCP and TCP applications. In: 2012 18th IEEE International Conference on Networks (ICON), pp. 41–45. IEEE (2012)Google Scholar
  2. 2.
    Qian, L., Carpenter, B.E.: Some observations on individual TCP flows behavior in network traffic traces. In: 2011 11th International Symposium on Communications and Information Technologies (ISCIT), pp. 354–359. IEEE (2011)Google Scholar
  3. 3.
    Carpenter, B.E.: Observed relationships between size measures of the internet. ACM SIGCOMM Comput. Commun. Rev. 39(2), 5–12 (2009)MathSciNetCrossRefGoogle Scholar
  4. 4.
    Lee, D.J., Carpenter, B.E., Brownlee, N.: Observations of UDP to TCP ratio and port numbers. In: 2010 Fifth International Conference on Internet Monitoring and Protection (ICIMP), pp. 99–104. IEEE (2010)Google Scholar
  5. 5.
    Nelson, R., Lawson, D., Lorier, P.: Analysis of long duration traces. ACM SIGCOMM Comput. Commun. Rev. 35(1), 45–52 (2005)CrossRefGoogle Scholar
  6. 6.
    Brownlee, N., Claffy, K.C.: Internet measurement. IEEE Internet Comput. 8(5), 30–33 (2004)CrossRefGoogle Scholar
  7. 7.
    Williamson, C.: Internet traffic measurement. IEEE Internet Comput. 5(6), 70–74 (2001)CrossRefGoogle Scholar
  8. 8.
    Arifler, D., De Veciana, G., Evans, B.L.: A factor analytic approach to inferring congestion sharing based on flow level measurements. IEEE/ACM Trans. Netw. (TON) 15(1), 67–79 (2007)CrossRefGoogle Scholar
  9. 9.
    Hwang, H., Yin, X., Wang, Z., et al.: The internet measurement of VoIP on different transport layer protocols. In: International Conference on Information Networking, ICOIN 2009, pp. 1–3. IEEE (2009)Google Scholar
  10. 10.
    Carpenter, B.E., Nichols, K.: Differentiated services in the Internet. Proc. IEEE 90(9), 1479–1494 (2002)CrossRefGoogle Scholar
  11. 11.
    Lee, D.J., Brownlee, N.: Passive measurement of one-way and two-way flow lifetimes. ACM SIGCOMM Comput. Commun. Rev. 37(3), 17–28 (2007). Some observations of Internet Stream LifetimesCrossRefGoogle Scholar
  12. 12.
    Brownlee, N.: Some observations of internet stream lifetimes. In: Dovrolis, C. (ed.) PAM 2005. LNCS, vol. 3431, pp. 265–277. Springer, Heidelberg (2005)CrossRefGoogle Scholar
  13. 13.
    Demichelis, C., Chimento, P.: IP packet delay variation metric for IP performance metrics (IPPM). RFC3393, IETF (2002)Google Scholar
  14. 14.
    jpcap network packet capture library.

Copyright information

© Springer-Verlag Berlin Heidelberg 2015

Authors and Affiliations

  1. 1.Institute of Information EngineeringChinese Academy of ScienceBeijingPeople’s Republic of China
  2. 2.Department of Computer ScienceThe University of AucklandAucklandNew Zealand
  3. 3.Department of ITSSEast China Normal UniversityShanghaiPeople’s Republic of China

Personalised recommendations