Markets and Negotiations in the Grid Economy: How decentralized and centralized economic approaches perform

Within several communities, dynamic and flexible allocation of computational resources is in the center of the current research interest. Grid Computing, e-Infrastructures, GÉANT, Utility Computing are several of the fields, which intensely work on such approaches. However, after seeing substantial progress on physically accessing computational facilities through innovative middleware (such as Globus), economic approaches have to be developed in order to efficiently utilize the resources. In this paper, the technical and economic performance of a centralized, market-based and a decentralized – Catallaxy-based – allocation mechanism for computational resources is compared using application-oriented metrics to measure the outcomes. The results show, that the efficiency of the mechanisms largely depends on the number of interacting stakeholders, the dynamics of the market as well as the topology applied for decentralized mechanisms