Multi-objective energy optimization in grid systems from a brain storming strategy

Nowadays, companies are more aware of an environmentally responsible use of computational resources. Terms like Green Computing promote energy savings in large-scale and distributed resource centers. Scheduling in distributed systems, as Grid Computing, is a challenging task in terms of time. Current research is considering energy savings as a new promising objective also for meta-schedulers. In this work, energy consumption and execution time are optimized simultaneously using a Multi-objective brain storm algorithm (MOBSA). This new algorithm is compared with two multi-objective algorithms: a novel algorithm based on the fireflies’ behavior—Multi-objective firefly algorithm (MO-FA)—and the well-known Non-dominated Sorting Genetic Algorithm (NSGA-II). Furthermore, other comparisons with real grid meta-schedulers such as Workload Management System from gLite, and Deadline Budget Constraint from Nimrod-G are carried out. The results show that MOBSA provides the best performance in any of the scenarios studied here.