Multiobjective Optimization Comparison - MOSWO vs MOGSA - for Solving the Job Scheduling Problem in Grid Environments

Scientists often have constraints from their experiments such as deadlines and budgets. For that reason, Job scheduling problem in Grid environments is not only important but also a challenging task. Both requirements - execution time and cost - are conflictive each other because faster resources usually involve higher costs. In this research, we compare two novel multiobjective algorithms from different fields - Complex Networks and Swarm approach - in an attempt to tackle the complex distributed infrastructure of Grid computing. On one hand, Multiobjective Small-World Optimization (MOSWO) is a multiobjective adaptation from algorithms based on the Small-World phenomenon, which is characteristic of complex scale-free networks. On the other hand, a novel swarm algorithm is the Multiobjective Gravitational Search Algorithm (MOGSA) inspired on gravitational attraction. Although both algorithms render good performance, MOGSA dominates in all the cases. Moreover, MOGSA attains improved performance with real schedulers such as the Workload Management System (WMS) from the most used European middleware gLite and the well-known Deadline Budget Constraint (DBC) algorithm from Nimrod-G.