Grid Meta Broker Selection Strategies for Job Reservation and Bidding

Grid computing is applying the resources of many computers in a network to a single problem at the same time. In scientific or technical problem it requires a great number of computer processing cycles or access to large amounts of data. In this paper we depict and evaluate broker selection strategies for job reservation and bidding. Especially, this paper analyzed two different types of existing algorithms simple and categorized aggregation algorithm. The first algorithm which aggregates the resource information acts as input for the categorized aggregation algorithm to assign rank for the resources. Meta broker allocates the job based on the rank. Form our assessment performed with simulation tool, we proposed advanced job reservation algorithm for resource allocation. Even though no resources are free to run the job, using this advanced resource algorithm we can reserve the resource for job allocation. In addition we proposed bidding technique when more than one users approach same resources. From the simulation results, we conclude that the proposed system reduces the execution time and generates better revenue for Meta broker.