Myconet: A fungi-inspired model for superpeer-based peer-to-peer overlay topologies

Paul L. Snyder, Rachel Greenstadt, Giuseppe Valetto

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

    Abstract

    Unstructured peer-to-peer networks can be extremely flexible, but, because of size, complexity, and high variability in peers' capacity and reliability, it is a continuing challenge to build peer-to-peer systems that are resilient to failure and effectively manage their available resources. We present Myconet, an approach to superpeer overlay construction inspired by the sophisticated, robust, root-like structures of fungal hyphae. Myconet models regular peers as biomass, and superpeers as hyphae that attract and concentrate biomass, while maintaining strong inter-connections with one another. Simulations of the Myconet peer-to-peer protocol show promising results in terms of network stabilization, response to catastrophic failure, capacity utilization, and proportion of peers to superpeers, when compared to other unstructured approaches.

    Original languageEnglish (US)
    Title of host publicationSASO 2009 - 3rd IEEE International Conference on Self-Adaptive and Self-Organizing Systems
    Pages40-50
    Number of pages11
    DOIs
    StatePublished - 2009
    EventSASO 2009 - 3rd IEEE International Conference on Self-Adaptive and Self-Organizing Systems - San Francisco, CA, United States
    Duration: Sep 14 2009Sep 18 2009

    Publication series

    NameSASO 2009 - 3rd IEEE International Conference on Self-Adaptive and Self-Organizing Systems

    Conference

    ConferenceSASO 2009 - 3rd IEEE International Conference on Self-Adaptive and Self-Organizing Systems
    Country/TerritoryUnited States
    CitySan Francisco, CA
    Period9/14/099/18/09

    ASJC Scopus subject areas

    • Computer Science Applications
    • Hardware and Architecture
    • Software

    Fingerprint

    Dive into the research topics of 'Myconet: A fungi-inspired model for superpeer-based peer-to-peer overlay topologies'. Together they form a unique fingerprint.

    Cite this