Skip to Main Content
Grid computing represents the main solution to integrate distributed and heterogeneous resources in global scale. However, the infrastructure necessary for maintaining a global grid in production is huge. Such fact has led to excessive power consumption. On the other hand, most green strategies for data centers are DVS (Dynamic Voltage Scaling)-based and become difficult to implement them in global grids. This paper proposes the HGreen heuristic (Heavier Tasks on Maximum Green Resource) and defines a workflow scheduling algorithm in order to implement it on global grids. HGreen algorithm aims to prioritize energy-efficient resources and explores workflow application profiles. Simulation results have shown that the proposed algorithm can significantly reduce the power consumption in global grids.
Date of Conference: 12-14 Dec. 2011