A simulation framework for studying economic resource management in grids

Abstract

Economic principles are increasingly being regarded as a way to address conflicting user requirements, to improve the effectiveness of grid resource management systems, and to deliver incentives for providers to join virtual organizations. Because economic resource management mechanisms can encourage grid participants to reveal the true valuations of their jobs and resources, the system becomes capable of making better scheduling decisions. A lot of exploratory research into different market mechanisms for grids is ongoing. Since it is impractical to conduct analysis of novel mechanisms on operational grids, most of this research is being carried out using simulation. This paper presents the Grid Economics Simulator (GES) in support of such research. The key design goals of the framework are enabling a wide variety of economic and non-economic forms of resource management while simultaneously supporting distributed execution of simulations and exhibiting good scalability properties. © 2008 Springer-Verlag Berlin Heidelberg.