An agent based semi-informed protocol for resource discovery in grids

Abstract. A Grid information system should rely upon two basic features: the replication and dissemination of information about Grid resources, and an intel-ligent logical distribution of such information among Grid hosts. This paper ex-amines an approach based on multi agent systems to build an information sys-tems in which metadata related to Grid resources is disseminated and logically organized according to a semantic classification of resources. Agents collect re-sources belonging to the same class in a restricted region of the Grid, so de-creasing the system entropy. A semi-informed resource discovery protocol ex-ploits the agents ’ work: query messages issued by clients are driven towards “representative peers ” which maintain information about a large number of re-sources having the required characteristics. Simulation analysis proves that the combined use of the resource mapping protocol (ARMAP) and the resource discovery protocol (ARDIP) allows users to find many useful results in a small amount of time.