Abstract: A grid job assignment scheme is proposed on the basis of the microeconomic method and swarm intelligence algorithm. With the grid resource management model developed and the classical auction model improved, the grid resource is priced via the improved auction procedure. An affinity degree function is introduced in order to seek for the optimal job assignment solution through such steps as antibody generation, selection, clone extension, immunological memory, self-adaptive mutation and adjustment, all are based on the improved immune optimization algorithm with both time and cost balanced in consideration. A GridSim-based simulation was implemented for the proposed scheme, and the results are compared with other two existing schemes. It follows that the performance of the proposed scheme is superior to the other two.