Soft error-aware energy-efficient task scheduling for workflow applications in DVFS-enabled cloud

Dynamic Voltage and Frequency Scaling (DVFS) has been widely used as a promising power management method to reduce the energy consumption of cloud workflows. However, due to the increasing chip density, lowering CPU voltages improperly in cloud data centers may inevitably increase soft error rate during workflow execution. Consequently, failures of timely completion of workflow applications may often take place, which raises serious concerns during the operation and maintenance of cloud data centers. To address such a problem, this paper proposes a soft error-aware energy-efficient task scheduling approach for workflow applications in DVFS-enabled cloud data centers. Under reliability and completion time constraints requested by tenants, our approach can generate energy-efficient task schedules for workflows by allocating tasks to appropriate virtual machines with specific operating frequencies. Comprehensive experiments on various well-known scientific workflow benchmarks show the effectivenss of our approach. Compared with state-of-the-art methods, our approach can achieve more than 30% energy savings while satisfying both reliability and completion time requirements.