Energy-Aware Scheduling of Real-Time Workflow Applications in Clouds Utilizing DVFS and Approximate Computations

As cloud services become more ubiquitous, green cloud computing attracts significant attention from both academia and industry. Towards this direction, in this paper we propose an energy-aware heuristic for the scheduling of real-time workflow applications in a cloud environment. Our approach utilizes per-core Dynamic Voltage and Frequency Scaling (DVFS) on the underlying heterogeneous multi-core processors and approximate computations, in order to fill in schedule gaps. Our goal is to provide timeliness and energy efficiency by trading off result precision, while keeping the average result precision of the completed jobs at an acceptable level. The proposed scheduling heuristic is compared to two other baseline policies. The simulation experiments reveal that our approach outperforms the other examined policies, providing promising results. To the best of our knowledge, such a technique that combines per-core DVFS and approximate computations in order to utilize schedule gaps in a virtualized environment with real-time workflow applications has never been discussed in the literature before.