Abstract
Autonomic server provisioning for performance assurance is a critical issue in Internet services. It is challenging to guarantee that requests flowing through a multi-tier system will experience an acceptable distribution of delays. The difficulty is mainly due to highly dynamic workloads, the complexity of underlying computer systems, and the lack of accurate performance models. We propose a novel autonomic server provisioning approach based on a model-independent self-adaptive Neural Fuzzy Control (NFC). Existing model-independent fuzzy controllers are designed manually on a trial-and-error basis, and are often ineffective in the face of highly dynamic workloads. NFC is a hybrid of control-theoretical and machine learning techniques. It is capable of self-constructing its structure and adapting its parameters through fast online learning. We further enhance NFC to compensate for the effect of server switching delays. Extensive simulations demonstrate that, compared to a rule-based fuzzy controller and a Proportional-Integral controller, the NFC-based approach delivers superior performance assurance in the face of highly dynamic workloads. It is robust to variation in workload intensity, characteristics, delay target, and server switching delays. We demonstrate the feasibility and performance of the NFC-based approach with a testbed implementation in virtualized blade servers hosting a multi-tier online auction benchmark.
- Abdelzaher, T. F., Shin, K. G., and Bhatti, N. 2002. Performance guarantees for web server end-systems: A control-theoretical approach. IEEE Trans. Parallel Distrib. Syst. 13, 1, 80--96. Google Scholar
Digital Library
- Amza, C., Chanda, A., Cox, A., Elnikety, S., Gil, R., Rajamani, K., Zwaenepoel, W., Cecchet, E., and Marguerite, J. 2002. Specification and implementation of dynamic web site benchmarks. In Proceedings of the IEEE International Workshop on Workload Characterization (WWC’02). 3--13.Google Scholar
- Bennani, M. N. and Menasce, D. A. 2005. Resource allocation for autonomic data centers using analytic performance models. In Proceedings of the IEEE International Conference on Autonomic Computing (ICAC’05). Google Scholar
Digital Library
- Bu, X., Rao, J., and Xu, C.-Z. 2009. A reinforcement learning approach to online web system auto-configuration. In Proceedings of the IEEE International Conference on Distributed Computing Systems (ICDCS’09). Google Scholar
Digital Library
- Chen, J., Soundararajan, G., and Amza, C. 2006. Autonomic provisioning of backend databases in dynamic content web servers. In Proceedings of the IEEE International Conference on Autonomic Computing (ICAC’06). Google Scholar
Digital Library
- Diao, Y., Hellerstein, J. L., Parekh, S., Shaihk, H., Surendra, M., and Tantawi, A. 2006. Modeling differentiated services of multi-tier web applications. In Proceedings of the IEEE International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS’06). Google Scholar
Digital Library
- Huebscher, M. C. and McCann, J. A. 2008. A survey of autonomic computing: Degrees, models, and applications. ACM Comput. Surv. 40, 3. Google Scholar
Digital Library
- Isci, C., Hanson, J. E., Whalley, I., Steinder, M., and Kephart, J. O. 2010. Runtime demand estimation for effective dynamic resource management. In Proceedings of the Network Operations and Management Symposium (NOMS’10).Google Scholar
- Jung, G., Hiltunen, M. A., Joshi, K. R., Schlichting, R. D., and Pu, C. 2010. Mistral: Dynamically managing power, performance, and adaptation cost in cloud infrastructures. In Proceedings of the IEEE International Conference on Distributed Computing Systems (ICDCS’10). Google Scholar
Digital Library
- Kamra, A., Misra, V., and Nahum, E. M. 2004. Yaksha: A self-tuning controller for managing the performance of 3-tiered web sites. In Proceedings of the IEEE International Workshop on Quality of Service (IWQoS’04).Google Scholar
- Karve, A., Kimbrel, T., Pacifici, G., Spreitzer, M., Steinder, M., Sviridenko, M., and Tantawi, A. 2006. Dynamic placement for clustered web applications. In Proceedings of the ACM International Conference on World Wide Web. Google Scholar
Digital Library
- Lama, P. and Zhou, X. 2009. Efficient server provisioning for end-to-end delay guarantee on multi-tier clusters. In Proceedings of the IEEE International Workshop on Quality of Service (IWQoS’09).Google Scholar
- Lama, P. and Zhou, X. 2010. Autonomic provisioning with self-adaptive neural fuzzy control for end-to-end delay guarantee. In Proceedings of the IEEE/ACM International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS’10). 151--160. Google Scholar
Digital Library
- Lama, P. and Zhou, X. 2012a. Efficient server provisioning with control for end-to-end delay guarantee on multi-tier clusters. IEEE Trans. Parall. Distrib. Syst. 23, 1, 78--86. Google Scholar
Digital Library
- Lama, P. and Zhou, X. 2012b. NINEPIN: Non-invasive and energy efficient performance isolation in virtualized servers. In Proceedings of the IEEE/IFIP Conference on Dependable Systems and Networks (DSN’12). 1--12. Google Scholar
Digital Library
- Leite, J. C. B., Kusic, D. M., Mosse, D., and Bertini, L. 2010. Stochastic approximation control of power and tardiness in a three-tier web-hosting cluster. In Proceedings of the IEEE International Conference on Autonomic Computing (ICAC’10). Google Scholar
Digital Library
- Lin, C. and Lee, C. S. G. 1992. Real-time supervised structure/parameter learning for fuzzy neural network. In Proceedings of the IEEE International Conference on Fuzzy Systems. 1283--1291.Google Scholar
- Lin, F.-J., Wai, R.-J., and Lee, C.-C. 1999. Fuzzy neural network position controller for ultrasonic motor drive using push-pull dc-dc converter. Control Theory Appl. 146, 1, 99--107.Google Scholar
Cross Ref
- Litoiu, M. 2007. A performance analysis method for autonomic computing systems. ACM Trans. Auton. Adapt. Syst. 2, 1. Google Scholar
Digital Library
- Liu, X., Sha, L., and Diao, Y. 2003. Online response time optimization of apache web server. In Proceedings of the International Workshop on Quality of Service (IWQoS’03). Google Scholar
Digital Library
- Liu, X., Heo, J., Sha, L., and Zhu, X. 2008. Queueing-model-based adaptive control of multi-tiered web applications. IEEE Trans. Netw. Service Manag. 5, 3, 157--167. Google Scholar
Digital Library
- Lu, C., Lu, Y., Abdelzaher, T. F., Stankovic, J. A., and Son, S. H. 2006. Feed back control architecture and design methodology for service delay guarantees in web servers. IEEE Trans. Parall. Distrib. Syst. 17, 9, 1014--1027. Google Scholar
Digital Library
- Meng, X., Isci, C., Kephart, J., Zhang, L., and Bouillet, E. 2010. Efficient resource provisioning in compute clouds via vm multiplexing. In Proceedings of the International Conference on Autonomic Computing (ICAC’10). Google Scholar
Digital Library
- Mi, N., Casale, G., Cherkasova, L., and Smirni, E. 2008. Burstiness in multi-tier applications: Symptoms, causes, and new models. In Proceedings of the ACM/IFIP/USENIX International Middleware Conference. Google Scholar
Digital Library
- Mi, N., Casale, G., Cherkasova, L., and Smirni, E. 2009. Injecting realistic burstiness to a traditional client-server benchmark. In Proceedings of the IEEE International Conference on Autonomic Computing (ICAC’09). Google Scholar
Digital Library
- Padala, P., Hou, K.-Y., Shin, K. G., Zhu, X., Uysal, M., Wang, Z., Singhal, S., and Merchant, A. 2009. Automated control of multiple virtualized resources. In Proceedings of the EuroSys Conference (EuroSys’09). 13--26. Google Scholar
Digital Library
- Rao, J. and Xu, C. 2011. Online capacity identification of multitier websites using hardware performance counters. IEEE Trans. Parall. Distrib. Syst. 22, 3, 426--438. Google Scholar
Digital Library
- RUBiS. 2013. Rice university bidding system. http://www.cs.rice.edu/CS/Systems/DynaServer/rubis.Google Scholar
- Sha, L., Liu, X., Lu, Y., and Abdelzaher, T. 2002. Queueing model based network server performance control. In Proceedings of the IEEE Real-Time Systems Symposium (RTSS’02). Google Scholar
Digital Library
- Singh, R., Sharma, U., Cecchet, E., and Shenoy, P. 2010. Autonomic mix-aware provisioning for non-stationary data center workloads. In Proceedings of the IEEE International Conference on Autonomic Computing (ICAC’10). 21--30. Google Scholar
Digital Library
- Stewart, C., Kelly, T., and Zhang, A. 2007. Exploiting nonstationarity for performance prediction. In Proceedings of the EuroSys Conference (EuroSys’07). 31--44. Google Scholar
Digital Library
- Tesauro, G., Jong, N. K., Das, R., and Bennani, M. N. 2006. A hybrid reinforcement learning approach to autonomic resource allocation. In Proceedings of the IEEE International Conference on Autonomic Computing (ICAC’06). Google Scholar
Digital Library
- Urgaonkar, B., Pacifici, G., Shenoy, P., Spreitzer, M., and Tantawi, A. 2005. An analytical model for multi-tier internet services and its applications. In Proceedings of the ACM SIGMETRICS International Conference on Measurement and Modeling of Computer Systems (SIGMETRICS’05). Google Scholar
Digital Library
- Urgaonkar, B., Shenoy, P., Chandra, A., Goyal, P., and Wood, T. 2008. Agile dynamic provisioning of multi-tier internet applications. ACM Trans. Auton. Adapt. Syst. 3, 1, 1--39. Google Scholar
Digital Library
- Villela, D., Pradhan, P., and Rubenstein, D. 2007. Provisioning servers in the application tier for e-commerce systems. ACM Trans. Internet Technol. 7, 1, 1--23. Google Scholar
Digital Library
- Wang, X. and Wang, Y. 2009. Co-con: Coordinated control of power and application performance for virtualized server clusters. In Proceedings of the IEEE International Workshop on Quality of Service (IWQoS’09).Google Scholar
- Watson, B. J., Marwah, M., Gmach, D., Chen, Y., Arlitt, M., and Wang, Z. 2010. Probabilistic performance modeling of virtualized resource allocation. In Proceedings of the IEEE International Conference on Autonomic Computing (ICAC’10). Google Scholar
Digital Library
- Wei, J. and Xu, C.-Z. 2006. eQoS: Provisioning of client-perceived end-to-end QoS guarantee in Web servers. IEEE Trans. Comput. 55, 12, 1543--1556. Google Scholar
Digital Library
- Welsh, M. and Culler, D. 2003. Adaptive overload control for busy Internet servers. In Proceedings of the 4th USENIX Symposium on Internet Technologies and Systems (USITS’03). Google Scholar
Digital Library
- Zhang, Q., Cherkasova, L., and Smirni, E. 2007. A regression-based analytic model for dynamic resource provisioning of multi-tier Internet applications. In Proceedings of the IEEE International Conference on Autonomic Computing (ICAC’07). Google Scholar
Digital Library
- Zhou, D. and Huang, W. W. 2009. Using a fuzzy classification approach to assess e-commerce web sites: An empirical investigation. ACM Trans. Internet Technol. 12, 9, 3. Google Scholar
Digital Library
- Zhou, X., Wei, J., and Xu, C.-Z. 2004. Processing rate allocation for proportional slowdown differentiation on Internet servers. In Proceedings of the IEEE International Parallel and Distributed Processing Symposium (IPDPS’04). 88--97.Google Scholar
Index Terms
Autonomic Provisioning with Self-Adaptive Neural Fuzzy Control for Percentile-Based Delay Guarantee
Recommendations
Autonomic Provisioning with Self-Adaptive Neural Fuzzy Control for End-to-end Delay Guarantee
MASCOTS '10: Proceedings of the 2010 IEEE International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication SystemsAutonomic server provisioning for performance assurance is a critical issue in data centers. It is important but challenging to guarantee an important performance metric, percentile-based end-to-end delay of requests flowing through a virtualized multi-...
Cache isolation and thin provisioning of hypervisor caches
LCN '12: Proceedings of the 2012 IEEE 37th Conference on Local Computer Networks (LCN 2012)Server virtualization has enabled resource consolidation and has minimized the need for additional and expensive hardware. Server virtualization has been widely deployed in a lot of organizations, because of the attractive benefits it offers like ...
Fuzzy Self-Adaptive PID Control Based on BP Neural Network for TCSC
IHMSC '09: Proceedings of the 2009 International Conference on Intelligent Human-Machine Systems and Cybernetics - Volume 01The flexible AC transmission system performance will be impacted directly by the controllers, therefore, the investigate in FACTS technology has been focus on the control system. With the characteristics and principles of the TCSC operation, this paper ...






Comments