A Self-Organising System for Resource Finding in Large-Scale Computational Grids

This paper presents a novel approach for resource finding and job allocation in a computational Grid. The proposed system is called HYGRA, for HYperspace-based Grid Resource Allocation. The basic principle involves the construction of a virtual hyperspace in which the available amount of each resource type is used as a geometric coordinate, making each Grid node representable as a point in this virtual hyperspace. A distributed overlay construction algorithm aims at connecting each node with the nearest k nodes w.r.t. the euclidean distance defined in the hyperspace. In this system, a job request, which can be also represented as a point, navigates the hyperspace, from node to node, following the overlay links which minimize the euclidean distance between the current node and the target point representing the job itself. The paper describes the algorithms for overlay construction and resource finding and assesses their validity and performances by means of a simulation approach.