A task scheduling method for energy-performance trade-off in clouds

Publication Type:
Conference Proceeding
Citation:
Proceedings - 18th IEEE International Conference on High Performance Computing and Communications, 14th IEEE International Conference on Smart City and 2nd IEEE International Conference on Data Science and Systems, HPCC/SmartCity/DSS 2016, 2017, pp. 1029 - 1036
Issue Date:
2017-01-20
Metrics:
Full metadata record
Files in This Item:
Filename Description Size
07828487.pdfPublished version367.14 kB
Adobe PDF
© 2016 IEEE. Energy consumption is an increasingly important issue for profit-driven cloud. Meanwhile, meeting users' expected response time is crucial for the real-time tasks. However, minimizing energy consumption may delay response time during task scheduling. Unfortunately, existing scheduling algorithms remain limit in balancing energy-saving and performance in elastic cloud environment. In view of this challenge, a task scheduling method for trade-off between energy and performance is proposed in this paper. Firstly, the energy-aware task scheduling problem in clouds is modeled by incorporating the virtualization technology. Then, an energy-aware resource provisioning algorithm is developed for generating available allocation plans for real time task by exploiting the elasticity of virtual resources. Finally, an energy-performance trade-off scheduling algorithm is designed for selecting final solution based on multi-objective optimization. The experimental evaluations are presented for validating our method.
Please use this identifier to cite or link to this item: