Testbeds and Research Infrastructure. Development of Networks and Communities. 7th International ICST Conference,TridentCom 2011, Shanghai, China, April 17-19, 2011, Revised Selected Papers

Research Article

Scalable Star-Topology Server-Array Based P2P Overlay Network Testbed

Download
334 downloads
  • @INPROCEEDINGS{10.1007/978-3-642-29273-6_14,
        author={Otso Kassinen and Erkki Harjula and Mika Ylianttila},
        title={Scalable Star-Topology Server-Array Based P2P Overlay Network Testbed},
        proceedings={Testbeds and Research Infrastructure. Development of Networks and Communities. 7th International ICST Conference,TridentCom 2011, Shanghai, China, April 17-19, 2011, Revised Selected Papers},
        proceedings_a={TRIDENTCOM},
        year={2012},
        month={5},
        keywords={Network simulation P2P overlay networks testbed control logic},
        doi={10.1007/978-3-642-29273-6_14}
    }
    
  • Otso Kassinen
    Erkki Harjula
    Mika Ylianttila
    Year: 2012
    Scalable Star-Topology Server-Array Based P2P Overlay Network Testbed
    TRIDENTCOM
    Springer
    DOI: 10.1007/978-3-642-29273-6_14
Otso Kassinen1,*, Erkki Harjula1,*, Mika Ylianttila1,*
  • 1: University of Oulu
*Contact email: Otso.Kassinen@ee.oulu.fi, Erkki.Harjula@ee.oulu.fi, Mika.Ylianttila@ee.oulu.fi

Abstract

We describe a scalable server-array based testbed for simulating various usage scenarios of peer-to-peer (P2P) overlay networks. Each server is responsible for a subset of the simulated peer processes, managed by the mechanisms presented herein. The system follows a star topology, where one master server acts as a point of control for a set of slave servers. We present both the structure of the system and the activities needed before, during, and after a simulation run in order to accomplish automated simulations, where each interesting combination of the variable parameters of the overlay network is evaluated. The functionality of the control scripts is explained in detail. Among other things, the system sets up the required start conditions for a P2P overlay simulation, manages the online-time and the specific P2P activities of each simulated peer, and facilitates the handling of the generated log files, from which the result statistics are derived.