Scheduling of Cloud Computing Tasks via Intelligent Optimization Methods

Abstract

Distributed green cloud datacenters (DGCDs) are increasingly deployed around the world. DGCDs integrate many renewable sources to provide clean power and decrease their operating cost. They are spread over multiple locations, where renewable energy availability, bandwidth prices and grid electricity costs have high geographical diversity. This paper focuses on delay-bounded applications in DGCDs and performs cost and energy-effective scheduling of multiple heterogeneous applications subject to delay-bound constraints. The minimization problem of operational cost of DGCDs is formulated and successfully solved by using Firefly, bat, and simulated annealing-bat algorithms. Data-driven experiments are conducted to assess and compare their effectiveness to solve it. The Firefly algorithm is shown to well outperform its peers.