Agent Based Economic Scheme for Seamless Job Scheduling

The bandwidth constraints, frequent disconnections and device mobility will affect the resource availability in a wireless grid. The device mobility from one cell to another has diverse impact on job execution. Due to limited resources and rational users, users may not be willing to share the resources. Scheduling of grid jobs in a complex, heterogeneous, resource constrained and dynamic wireless grid environment is challenging. This paper proposes a scheme for seamless job scheduling by using software agents in bandwidth con-strained wireless grids. An economic scheme by using non-cooperative bargaining game is designed to encourage resource sharing depending upon grid market dynamics. The scheme consists of two agencies: grid information service and resource broker agency. The agents in agencies interact and cooperate with each other to discover reliable devices, negotiate resource cost and bandwidth, and support seamless job execution. Based on the mobility of a device, agents decide dynamically either to continue or terminate execution of a sched-uled job. The scheme is simulated to evaluate the performance parameters such as expected surplus, job completion time and job execution rate. We observed that the proposed scheme performs better than the existing work.