Resource Matching Based on Ontology and Mobile Agent in Grid Environments

Grid computing environments can share resources and services in a large-scale. It allows every user to extract aggregated computational power from idle PCs in the Internet, thereby to accomplish high performance computing task with low cost. However, the use of the grid resource will be a complex task for an ordinary user. To improve the resource matching (resource search and selection), we proposed a resource matching scheme. This scheme had the resource specification description based on ontology; used mobile Agent as the delegate for the resource ontology of the provider/consumer and matched resource by the market broker. The advantages of this scheme are: (1) the use of the grid resource becomes very easy for an ordinary user. (2) Resource prices are dynamically decided on market principle. (3) The resource specification is described by ontology conception (no constraint condition), so that the scheme has expansibility.