Diagonal tuple space search in two dimensions

Mikko Alutoin, Pertti Raatikainen

    Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

    2 Citations (Scopus)

    Abstract

    Due to the evolution of the Internet and its services, the process of forwarding packets in routers is becoming more complex. In order to execute the sophisticated routing logic of modern firewalls, multidimensional packet classification is required. Unfortunately, the multidimensional packet classification algorithms are known to be either time or storage hungry in the general case. It has been anticipated that more feasible algorithms could be obtained for conflict-free classifiers. This paper proposes a novel two-dimensional packet classification algorithm applicable to the conflict-free classifiers. It derives from the well-known tuple space paradigm and it has the search cost of Ο(log w) and storage complexity of Ο(n2w log w), where w is the width of the protocol fields given in bits and n is the number of rules in the classifier. This is remarkable because without the conflict-free constraint the search cost in the two-dimensional tuple space is Θ(w).
    Original languageEnglish
    Title of host publicationNetworking 2004
    PublisherSpringer
    Pages308-319
    ISBN (Electronic)978-3-540-24693-0
    ISBN (Print)978-3-540-21959-0
    DOIs
    Publication statusPublished - 2004
    MoE publication typeA4 Article in a conference publication
    Event3rd International IFIP-TC6 Networking Conference, Networking 2004 - Athens, Greece
    Duration: 9 May 200414 May 2004

    Publication series

    SeriesLecture Notes in Computer Science
    Volume3042
    ISSN0302-9743

    Conference

    Conference3rd International IFIP-TC6 Networking Conference, Networking 2004
    Country/TerritoryGreece
    CityAthens
    Period9/05/0414/05/04

    Keywords

    • routing
    • packet classification
    • algorithms

    Fingerprint

    Dive into the research topics of 'Diagonal tuple space search in two dimensions'. Together they form a unique fingerprint.

    Cite this