4th International ICST Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services

Research Article

Distributed Service Deployment in Mobile Ad-Hoc Networks

  • @INPROCEEDINGS{10.1109/MOBIQ.2007.4451002,
        author={Artin Avanes and Johann-Christoph Freytag and Christof Bornh\o{}vd},
        title={Distributed Service Deployment in Mobile Ad-Hoc Networks},
        proceedings={4th International ICST Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services},
        publisher={IEEE},
        proceedings_a={MOBIQUITOUS},
        year={2008},
        month={2},
        keywords={Ad hoc networks  Delay  Information retrieval  Intelligent sensors  Personal digital assistants  Prototypes  Real time systems  Scalability  Simple object access protocol  Smart phones},
        doi={10.1109/MOBIQ.2007.4451002}
    }
    
  • Artin Avanes
    Johann-Christoph Freytag
    Christof Bornhövd
    Year: 2008
    Distributed Service Deployment in Mobile Ad-Hoc Networks
    MOBIQUITOUS
    IEEE
    DOI: 10.1109/MOBIQ.2007.4451002
Artin Avanes1,*, Johann-Christoph Freytag1,*, Christof Bornhövd2,*
  • 1: Humboldt-Universitat zu Berlin 10099 Berlin, Germany
  • 2: SAP Labs, LLC Palo Alto, 94304 California
*Contact email: avanes@informatik.hu-berlin.de, freytag@informatik.hu-berlin.de, christof.bornhoevd@sap.com

Abstract

Today’s applications are increasingly composed out of services. Standardized protocols, such as SOAP, WSDL, and UDDI are used to discover and invoke remotely located services. Nowadays, improved resource capabilities of mobile devices, e.g. PDAs, smart phones, and sensor devices allow the execution of services even on these smaller computing devices. In comparison to traditional centralized process management, a decentralized, cooperative execution of services on embedded real-time systems leads to higher system scalability, better system response time and higher data accuracy. In this paper we describe an efficient way to deploy services onto highly distributed, mobile, and unreliable devices. To achieve an efficient resource tracking we utilize different group-based data retrieval strategies. Furthermore, we present a prototype system that implements our distributed service deployment algorithm and that evaluates our approach in terms of scalability for different network topologies.