2nd International ICST Workshop on OMNeT++

Research Article

Exchangeable, application-independent load balancing for P2P simulation frameworks

  • @INPROCEEDINGS{10.4108/ICST.SIMUTOOLS2009.5567,
        author={Daniel  Warneke and Ulf  Rerrer-Brusch},
        title={Exchangeable, application-independent load balancing for P2P simulation frameworks},
        proceedings={2nd International ICST Workshop on OMNeT++},
        publisher={ACM},
        proceedings_a={OMNET++},
        year={2010},
        month={5},
        keywords={Peer-To-Peer Simulation Load Balancing Modularization},
        doi={10.4108/ICST.SIMUTOOLS2009.5567}
    }
    
  • Daniel Warneke
    Ulf Rerrer-Brusch
    Year: 2010
    Exchangeable, application-independent load balancing for P2P simulation frameworks
    OMNET++
    ICST
    DOI: 10.4108/ICST.SIMUTOOLS2009.5567
Daniel Warneke1,*, Ulf Rerrer-Brusch1,*
  • 1: Technische Universität Berlin, Berlin, Germany.
*Contact email: warneke@cs.tu-berlin.de, urerrer@cs.tu-berlin.de

Abstract

Today many peer-to-peer (P2P) simulation frameworks feature a variety of recent years' research outcomes as modular building blocks, allowing others to easily reuse these blocks in further simulations and approach more advanced issues more rapidly. However, the efforts in the field of P2P load balancing have been excluded from that development so far, as the proposed techniques often impose too many dependencies between the load balancing algorithms and the application to be put in loosely-coupled components. This paper discusses how load balancing algorithms that rely on the virtual server concept can be separated from the application and run in a modular container with a unified communication interface. We discuss design fundamentals for such a load balancing container based on a variety of existing load balancing techniques and present our implementation for the OverSim framework. With our work load balancing becomes a reusable building block for P2P applications which contributes to the process of building rich and modular simulation environments.