On the integration of economic principles in grid resource management

On the integration of economic principles in grid resource management

Abstract

Grid computing has emerged as a paradigm for integrating and sharing IT resources and services across administrative and organizational boundaries. Large-scale grids such as those realized by the European “Enabling Grids for E-Science (EGEE)” project and the TeraGrid project in the US, now integrate compute and storage resources hosted by hundreds of research institutions across the globe. Grid technology allows researchers to tackle computational problems on an unprecedented scale. In addition, the technology delivers a platform for scientific collaboration and allows for more efficient use of resources by integrating and sharing them on a large scale. However, the developments in grid technology have given limited consideration to the realization of a model in which resources are shared among users and providers that have a potentially loose prefatory relationship.

As a consequence, mechanisms are at present missing that deliver clear incentives to resource owners to share their resources openly and for users to use resources in a well-considered manner. In addition, a flexible and efficient mechanism is currently missing that determines who obtains access to resources and services on a grid, at what time, and at what cost. As a result, the full potential of grid technology is not realized and many opportunities for increasing the efficiency of these systems remain.