Profit-Aware Spatial Task Scheduling in Distributed Green Clouds
Haitao Yuan, Jing Cun Bi · 2019
More and more large-scale enterprises choose distributed green clouds (DGCs) to cost-effectively deploy their applications. The significant increase of users' tasks makes it highly challenging to achieve profit maximization for a DGC provider under the fact that prices of power grid, revenues, and the amount of wind and solar energy in DGCs all change with different sites. This work develops a Profit-Aware Spatial Task Scheduling (PASTS) method for the profit maximization of a DGC provider. PASTS well investigates such spatial differences of these mentioned factors, and it smartly schedules tasks to meet their response time constraints. A nonlinear constrained program is designed and tackled by a hybrid meta-heuristic algorithm that combines particle swarm optimization with genetic mechanism and simulated annealing. Realistic data-based results prove that PASTS provides higher profit and throughput than two recent typical algorithms.