A Resource Discovery Method of Grid Based on Resource Classification

Abstract

Grid is a 'virtual supercomputer' which combines the resources distributed in different palaces through the network. Aiming at the continual increase of grid resource, and on the basis of comparative analysis of the resource discovery mechanism of the existing networking, a mixed resource discovery model, which combines the distributing and centralizing mechanism together, is proposed in this paper. This model not only satisfies the grid systempsilas extendibility, but also reduces the massive network communication under the distributed mechanism. By taking use of combining centralized and distributed way, and classifying resources and P2P technology, this model is of better effect on grid resource extension and stability of the system.