Job-Scheduling with Resource Availability Prediction for Volunteer-Based Grid Computing

In a volunteer-based grid computing environment, one big challenge for effective job allocation is resource availability. As resources in this environment are volatile and may become frequently unavailable, matching guest jobs to suitable resources is very important. To improve scheduling performance in such an unreliable computing environment, especially in terms of avoiding job completion failure due to resource unavailability, this paper proposes a new jobscheduling algorithm. This scheduling algorithm is based on an existing resource availability prediction technique that anticipates future availability of resources to help make better job allocation decisions. Simulation results shows this new job scheduling algorithm provides better results in terms of ensuring jobs are processed successfully.