The Community for Technology Leaders
RSS Icon
Issue No.04 - April (2013 vol.24)
pp: 681-690
Xiangping Bu , Dept. of Electr. & Comput. Eng., Wayne State Univ., Detroit, MI, USA
Jia Rao , Dept. of Electr. & Comput. Eng., Wayne State Univ., Detroit, MI, USA
Cheng-Zhong Xu , Dept. of Electr. & Comput. Eng., Wayne State Univ., Detroit, MI, USA
Cloud computing has a key requirement for resource configuration in a real-time manner. In such virtualized environments, both virtual machines (VMs) and hosted applications need to be configured on-the-fly to adapt to system dynamics. The interplay between the layers of VMs and applications further complicates the problem of cloud configuration. Independent tuning of each aspect may not lead to optimal system wide performance. In this paper, we propose a framework, namely CoTuner, for coordinated configuration of VMs and resident applications. At the heart of the framework is a model-free hybrid reinforcement learning (RL) approach, which combines the advantages of Simplex method and RL method and is further enhanced by the use of system knowledge guided exploration policies. Experimental results on Xen-based virtualized environments with TPC-W and TPC-C benchmarks demonstrate that CoTuner is able to drive a virtual server cluster into an optimal or near-optimal configuration state on the fly, in response to the change of workload. It improves the systems throughput by more than 30 percent over independent tuning strategies. In comparison with the coordinated tuning strategies based on basic RL or Simplex algorithm, the hybrid RL algorithm gains 25 to 40 percent throughput improvement.
virtualisation, cloud computing, learning (artificial intelligence), real-time systems, resource allocation, virtual machines, coordinated tuning strategies, coordinated self-configuration, virtual machines, cloud computing, real-time resource configuration, VM, system dynamics, cloud configuration, CoTuner framework, model-free hybrid reinforcement learning, model-free hybrid RL approach, Simplex method, RL method, system knowledge, Xen-based virtualized environments, TPC-W benchmarks, TPC-C benchmarks, virtual server cluster, near-optimal configuration state, system throughput improvement, Servers, Tuning, System performance, Throughput, Resource management, Clustering algorithms, Learning, autonomic configuration, Cloud computing, reinforcement learning
Xiangping Bu, Jia Rao, Cheng-Zhong Xu, "Coordinated Self-Configuration of Virtual Machines and Appliances Using a Model-Free Learning Approach", IEEE Transactions on Parallel & Distributed Systems, vol.24, no. 4, pp. 681-690, April 2013, doi:10.1109/TPDS.2012.174
[1] D. Ongaro, A.L. Cox, and S. Rixner, "Scheduling i/o in Virtual Machine Monitors," Proc. Fourth ACM SIGPLAN/SIGOPS Int'l Conf. Virtual Execution Environments (VEE), 2008.
[2] D. Gupta, L. Cherkasova, R. Gardner, and A. Vahdat, "Enforcing Performance Isolation across Virtual Machines in Xen," Middleware: Proc. ACM/IFIP/USENIX Int'l Conf. Middleware , 2006.
[3] J. Rao, X. Bu, C.-Z. Xu, L. Wang, and G. Yin, "VCONF: A Reinforcement Learning Approach to Virtual Machines Auto-Configuration," Proc. Sixth Int'l Conf. Autonomic Computing (ICAC), 2009.
[4] A. Verma, P. De, V. Mann, T.K. Nayak, A. Purohit, G. Dasgupta, and R. Kothari, "Brownmap: Enforcing Power Budget in Shared Data Centers," Middleware: Proc. Int'l Conf. Distributed Systems Platforms and Open Distributed Processing /Open Distributed Processing, pp. 42-63, 2010.
[5] R. Nathuji, A. Kansal, and A. Ghaffarkhah, "Q-Clouds: Managing Performance Interference Effects for Qos-Aware Clouds," Proc. European Conf. Computer Systems (EuroSys), pp. 237-250, 2010.
[6] P. Padala, K.G. Shin, X. Zhu, M. Uysal, Z. Wang, S. Singhal, A. Merchant, and K. Salem, "Adaptive Control of Virtualized Resources in Utility Computing Environments," Proc. ACM Second SIGOPS/EuroSys European Conf. Computer Systems (EuroSys), 2007.
[7] P. Padala, K.-Y. Hou, K.G. Shin, X. Zhu, M. Uysal, Z. Wang, S. Singhal, and A. Merchant, "Automated Control of Multiple Virtualized Resources," Proc. European Conf. Computer Systems (EuroSys), pp. 13-26, 2009.
[8] Y. Wang, X. Wang, M. Chen, and X. Zhu, "Power-Efficient Response Time Guarantees for Virtualized Enterprise Servers," Proc. IEEE Real-Time Systems Symp., pp. 303-312, 2008.
[9] Y. Wang, R. Deaver, and X. Wang, "Virtual Batching: Request Batching for Energy Conservation in Virtualized Servers," Proc. 18th Int'l Workshop Quality of Service (IWQoS), 2010.
[10] X. Liu, L. Sha, Y. Diao, S. Froehlich, J.L. Hellerstein, and S.S. Parekh, "Online Response time Optimization of Apache Web Server," Proc. 11th Int'l Workshop Quality of Service (IWQoS), pp. 461-478, 2003.
[11] B. Xi, Z. Liu, M. Raghavachari, C.H. Xia, and L. Zhang, "A Smart Hill-Climbing Algorithm for Application Server Configuration," Proc. 13th Int'l Conf. World Wide Web (WWW), pp. 287-296, 2004.
[12] Y. Zhang, W. Qu, and A. Liu, "Automatic Performance Tuning for J2EE Application Server Systems," Proc. Sixth Int'l Conf. Web Information Systems Eng. (WISE), 2005.
[13] I.-H. Chung and J.K. Hollingsworth, "Automated Cluster-Based Web Service Performance Tuning," Proc. IEEE 13th Int'l Symp. High Performance Distributed (HPDC), pp. 36-44, 2004.
[14] W. Zheng, R. Bianchini, and T.D. Nguyen, "Automatic Configuration of Internet Services," Proc. Second ACM SIGOPS/EuroSys European Conf. Computer Systems (EuroSys), 2007.
[15] R.S. Sutton and A.G. Barto, Reinforcement Learning: An Introduction. MIT Press, 1998.
[16] X. Bu, J. Rao, and C.-Z. Xu, "A Reinforcement Learning Approach to Online Web Systems Auto-Configuration," Proc. IEEE 29th Int'l Conf. Distributed Computing Systems (ICDCS), 2009.
[17] http://www.tpc.orgtpcw, 2012.
[18] http://www.tpc.orgtpcc, 2012.
[19] A. Whitaker, R.S. Cox, and S.D. Gribble, "Configuration Debugging as Search: Finding the Needle in the Haystack," Proc. Sixth Conf. Symp. Operating Systems Design & Implementation (OSDI), 2004.
[20] E. Kwan, S. Lightstone, K.B. Schiefer, A.J. Storm, and L. Wu, "Automatic Database Configuration for DB2 Universal Database: Compressing Years of Performance Expertise into Seconds of Execution," Proc. Büro, Technik und Wissenschaft BTW, pp. 620-629, 2003.
[21] J. Wei and C.-Z. Xu, "eQoS: Provisioning of Client-Perceived End-to-End QoS Guarantees in Web Servers," IEEE Trans. Computers, vol. 55, no. 12, pp. 1543-1556, Dec. 2006.
[22] C.-Z. Xu, J. Wei, and F. Liu, "Model Predictive Feedback Control for End-to-End QoS Guarantees in Web Servers," Computer, vol. 41, no. 3, pp. 66-72, Mar. 2008.
[23] J. Gong and C.-Z. Xu, "A Gray-Box Feedback Control Approach for System-Level Peak Power Management," Proc. Int'l Conf. Parallel Processing (ICPP), 2010.
[24] J. Gong and C.-Z. Xu, "vPnP: Automated Coordination of Power and Performance in Virtualized Datacenters," Proc. Int'l Workshop Quality of Service (IWQoS), 2010.
[25] G. Tesauro, "Online Resource Allocation Using Decompositional Reinforcement Learning," Proc. 20th Nat'l Conf. Artificial Intelligence (AAAI), 2005.
[26] G. Tesauro, N.K. Jong, R. Das, and M.N. Bennani, "On the use of Hybrid Reinforcement Learning for Autonomic Resource Allocation," Cluster Computing, vol. 10, pp. 287-299, 2007.
[27] G. Tesauro, R. Das, H. Chan, J. Kephart, D. Levine, F. Rawson, and C. Lefurgy, "Managing Power Consumption and Performance of Computing Systems Using Reinforcement Learning," Proc. Advances in Neural Information Processing Systems 20, 2007.
[28] A. Bar-Hillel, A. Di-Nur, L. Ein-Dor, R. Gilad-Bachrach, and Y. Ittach, "Workstation Capacity Tuning Using Reinforcement Learning," Proc. ACM/IEEE Int'l Conf. Supercomputing (SC), 2007.
76 ms
(Ver 2.0)

Marketing Automation Platform Marketing Automation Tool