Deep Reinforcement Learning for Multi-resource Cloud Job Scheduling
The resource scheduling problem in the cloud environment has always been a difficult and hot research field of cloud computing. The difficult problem of online decision-making tasks for resource management in a complex cloud environment can be solved by combining the excellent decision-making ability of reinforcement learning and the strong environmental awareness ability of deep learning. This paper proposes a multi-resource cloud job scheduling strategy in cloud environment based on Deep Q-network algorithm to minimize the average job completion time and average job slowdown. The experimental results show that the scheduling strategy is better than the scheduling strategy based on the standard policy gradient algorithm, and accelerate the convergence speed.
KeywordsCloud computing Deep reinforcement learning Job scheduling
The work presented in this paper was supported by National Natural Science Foundation of China (61772145, 61672174).
- 1.Wang, T., Liu, Z., Chen, Y., Xu, Y., Dai, X.: Load balancing task scheduling based on genetic algorithm in cloud computing. In: Proceedings of the 12th International Conference on Dependable, Autonomic and Secure Computing, pp. 146–152 (2014)Google Scholar
- 4.Sutton, R.S., Barto, A.G.: Reinforcement Learning: An Introduction. MIT Press, Cambridge (1998)Google Scholar
- 5.Dutreilh, X., Kirgizov, S., Melekhova, O.: Using reinforcement learning for autonomic resource allocation in clouds: towards a fully automated workflow, pp. 67–74 (2011)Google Scholar
- 7.Galstyan, A., Czajkowski, K., Lerman, K.: Resource allocation in the grid using reinforcement learning. In: International Joint Conference on Autonomous Agents and Multiagent Systems, pp. 1314–1315 (2004)Google Scholar
- 9.Peng, Z., Cui, D., Zuo, J.: Research on cloud computing resources provisioning based on reinforcement learning. Math. Prob. Eng. 2015, 1–12 (2015)Google Scholar
- 10.Peng, Z., Cui, D., Ma, Y., Xiong, J., Xu, B., Lin, W.: A reinforcement learning-based mixed job scheduler scheme for cloud computing under SLA constraint. In: International Conference on Cyber Security and Cloud Computing, pp. 142–147 (2016)Google Scholar
- 12.Mao, H., Alizadeh, M., Menache, I.: Resource management with deep reinforcement learning. In: ACM Workshop on Hot Topics in Networks, pp. 50–56 (2016)Google Scholar
- 13.Mnih, V., Kavukcuoglu, K., Silver, D.: Playing Atari with deep reinforcement learning. Computer Science (2013)Google Scholar
- 14.Hinton, G.: Overview of mini-batch gradient descent. Neural Networks for Machine Learning. https://www.coursera.org/learn/neural-networks. Accessed 13 June 2018
- 15.Schulman, J., Levine, S., Moritz, P.: Trust region policy optimization. In: Computer Science, pp. 1889–1897 (2015)Google Scholar
- 17.Liu, Q., Zhai, J.W., Zhang, Z.Z.: A survey on deep reinforcement learning. Chin. J. Comput. 40, 1–28 (2018)Google Scholar