8th IEEE International Conference on Collaborative Computing: Networking, Applications and Worksharing

Research Article

Incentive Mechanisms for Opportunistic Cloud Computing Services

Download491 downloads
  • @INPROCEEDINGS{10.4108/icst.collaboratecom.2012.250403,
        author={Eric Kuada and Henning Olesen},
        title={Incentive Mechanisms for Opportunistic Cloud Computing Services},
        proceedings={8th IEEE International Conference on Collaborative Computing: Networking, Applications and Worksharing},
        publisher={IEEE},
        proceedings_a={COLLABORATECOM},
        year={2012},
        month={12},
        keywords={game theory mechanism design opportunistic cloud computing services},
        doi={10.4108/icst.collaboratecom.2012.250403}
    }
    
  • Eric Kuada
    Henning Olesen
    Year: 2012
    Incentive Mechanisms for Opportunistic Cloud Computing Services
    COLLABORATECOM
    ICST
    DOI: 10.4108/icst.collaboratecom.2012.250403
Eric Kuada1,*, Henning Olesen1
  • 1: Aalborg University
*Contact email: kuada@cmi.aau.dk

Abstract

Opportunistic Cloud Computing Service (OCCS) is a social network approach to the provisioning and management of cloud computing services for enterprises. The OCCS network may suffer from the free riding problem where members are selfish and will only want to use services on the platform without ever contributing resources. It may also suffer from resource wastage from members or external entities trying to attack the system so that genuine users are deprived of valuable resources. The purpose of this paper is to design incentive schemes that will encourage the contribution of resources to the OCCS platform as well as the efficient usage of these resources. We employ game theory and mechanism design to model and design the incentive schemes. We present two game models and show the existence of a pure strategy Nash equilibrium for both the cooperative and non-cooperative games. Three base incentive schemes are presented and two advanced schemes one based on discount factor and the other a stochastic scheme are also presented. We perform analytical evaluation of our incentive schemes and conclude that the schemes meet the desired properties of budget-balance, ex-post individual rationality, incentive compatibility, allocative efficiency, robustness, and flexible to accommodate changing user behavior on the platform.