Identifying elephant flows through periodically sampled packets

Tatsuya Mori, Masato Uchida, Ryoichi Kawahara, Jianping Pan, Shigeki Goto

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    115 Citations (Scopus)

    Abstract

    Identifying elephant flows is very important in developing effective and efficient traffic engineering schemes. In addition, obtaining the statistics of these flows is also very useful for network operation and management. On the other hand, with the rapid growth of link speed in recent years, packet sampling has become a very attractive and scalable means to measure flow statistics; however, it also makes identifying elephant flows become much more difficult. Based on Bayes' theorem, this paper develops techniques and schemes to identify elephant flows in periodically sampled packets. We show that our basic framework is very flexible in making appropriate trade-offs between false positives (misidentified flows) and false negatives (missed elephant flows) with regard to a given sampling frequency. We further validate and evaluate our approach by using some publicly available traces. Our schemes are generic and require no per-packet processing; hence, they allow a very cost-effective implementation for being deployed in large-scale high-speed networks.

    Original languageEnglish
    Title of host publicationProceedings of the 2004 ACM SIGCOMM Internet Measurement Conference, IMC 2004
    Pages115-120
    Number of pages6
    Publication statusPublished - 2004
    EventProceedings of the 2004 ACM SIGCOMM Internet Measurement Conference, IMC 2004 - Taormina
    Duration: 2004 Oct 252004 Oct 27

    Other

    OtherProceedings of the 2004 ACM SIGCOMM Internet Measurement Conference, IMC 2004
    CityTaormina
    Period04/10/2504/10/27

    Fingerprint

    Statistics
    Sampling
    HIgh speed networks
    Processing
    Costs

    Keywords

    • Bayes' theorem
    • Flow statistics
    • Measurement
    • Packet sampling
    • The elephant and mice phenomenon

    ASJC Scopus subject areas

    • Engineering(all)

    Cite this

    Mori, T., Uchida, M., Kawahara, R., Pan, J., & Goto, S. (2004). Identifying elephant flows through periodically sampled packets. In Proceedings of the 2004 ACM SIGCOMM Internet Measurement Conference, IMC 2004 (pp. 115-120)

    Identifying elephant flows through periodically sampled packets. / Mori, Tatsuya; Uchida, Masato; Kawahara, Ryoichi; Pan, Jianping; Goto, Shigeki.

    Proceedings of the 2004 ACM SIGCOMM Internet Measurement Conference, IMC 2004. 2004. p. 115-120.

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Mori, T, Uchida, M, Kawahara, R, Pan, J & Goto, S 2004, Identifying elephant flows through periodically sampled packets. in Proceedings of the 2004 ACM SIGCOMM Internet Measurement Conference, IMC 2004. pp. 115-120, Proceedings of the 2004 ACM SIGCOMM Internet Measurement Conference, IMC 2004, Taormina, 04/10/25.
    Mori T, Uchida M, Kawahara R, Pan J, Goto S. Identifying elephant flows through periodically sampled packets. In Proceedings of the 2004 ACM SIGCOMM Internet Measurement Conference, IMC 2004. 2004. p. 115-120
    Mori, Tatsuya ; Uchida, Masato ; Kawahara, Ryoichi ; Pan, Jianping ; Goto, Shigeki. / Identifying elephant flows through periodically sampled packets. Proceedings of the 2004 ACM SIGCOMM Internet Measurement Conference, IMC 2004. 2004. pp. 115-120
    @inproceedings{5b7a924f3e5b47b1929a7004db2af188,
    title = "Identifying elephant flows through periodically sampled packets",
    abstract = "Identifying elephant flows is very important in developing effective and efficient traffic engineering schemes. In addition, obtaining the statistics of these flows is also very useful for network operation and management. On the other hand, with the rapid growth of link speed in recent years, packet sampling has become a very attractive and scalable means to measure flow statistics; however, it also makes identifying elephant flows become much more difficult. Based on Bayes' theorem, this paper develops techniques and schemes to identify elephant flows in periodically sampled packets. We show that our basic framework is very flexible in making appropriate trade-offs between false positives (misidentified flows) and false negatives (missed elephant flows) with regard to a given sampling frequency. We further validate and evaluate our approach by using some publicly available traces. Our schemes are generic and require no per-packet processing; hence, they allow a very cost-effective implementation for being deployed in large-scale high-speed networks.",
    keywords = "Bayes' theorem, Flow statistics, Measurement, Packet sampling, The elephant and mice phenomenon",
    author = "Tatsuya Mori and Masato Uchida and Ryoichi Kawahara and Jianping Pan and Shigeki Goto",
    year = "2004",
    language = "English",
    isbn = "1581138210",
    pages = "115--120",
    booktitle = "Proceedings of the 2004 ACM SIGCOMM Internet Measurement Conference, IMC 2004",

    }

    TY - GEN

    T1 - Identifying elephant flows through periodically sampled packets

    AU - Mori, Tatsuya

    AU - Uchida, Masato

    AU - Kawahara, Ryoichi

    AU - Pan, Jianping

    AU - Goto, Shigeki

    PY - 2004

    Y1 - 2004

    N2 - Identifying elephant flows is very important in developing effective and efficient traffic engineering schemes. In addition, obtaining the statistics of these flows is also very useful for network operation and management. On the other hand, with the rapid growth of link speed in recent years, packet sampling has become a very attractive and scalable means to measure flow statistics; however, it also makes identifying elephant flows become much more difficult. Based on Bayes' theorem, this paper develops techniques and schemes to identify elephant flows in periodically sampled packets. We show that our basic framework is very flexible in making appropriate trade-offs between false positives (misidentified flows) and false negatives (missed elephant flows) with regard to a given sampling frequency. We further validate and evaluate our approach by using some publicly available traces. Our schemes are generic and require no per-packet processing; hence, they allow a very cost-effective implementation for being deployed in large-scale high-speed networks.

    AB - Identifying elephant flows is very important in developing effective and efficient traffic engineering schemes. In addition, obtaining the statistics of these flows is also very useful for network operation and management. On the other hand, with the rapid growth of link speed in recent years, packet sampling has become a very attractive and scalable means to measure flow statistics; however, it also makes identifying elephant flows become much more difficult. Based on Bayes' theorem, this paper develops techniques and schemes to identify elephant flows in periodically sampled packets. We show that our basic framework is very flexible in making appropriate trade-offs between false positives (misidentified flows) and false negatives (missed elephant flows) with regard to a given sampling frequency. We further validate and evaluate our approach by using some publicly available traces. Our schemes are generic and require no per-packet processing; hence, they allow a very cost-effective implementation for being deployed in large-scale high-speed networks.

    KW - Bayes' theorem

    KW - Flow statistics

    KW - Measurement

    KW - Packet sampling

    KW - The elephant and mice phenomenon

    UR - http://www.scopus.com/inward/record.url?scp=14944376803&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=14944376803&partnerID=8YFLogxK

    M3 - Conference contribution

    AN - SCOPUS:14944376803

    SN - 1581138210

    SN - 9781581138214

    SP - 115

    EP - 120

    BT - Proceedings of the 2004 ACM SIGCOMM Internet Measurement Conference, IMC 2004

    ER -