Abstract
The capacity needs of online services are mainly determined by the volume of user loads. For large-scale distributed systems running such services, it is quite difficult to match the capacities of various system components. In this paper, a novel and systematic approach is proposed to profile services for resource optimization and capacity planning. We collect resource consumption related measurements from various components across distributed systems and further search for constant relationships between these measurements. If such relationships always hold under various workloads along time, we consider them as invariants of the underlying system. After extracting many invariants from the system, given any volume of user loads, we can follow these invariant relationships sequentially to estimate the capacity needs of individual components. By comparing the current resource configurations against the estimated capacity needs, we can discover the weakest points that may deteriorate system performance. Operators can consult such analytical results to optimize resource assignments and remove potential performance bottlenecks. In this paper, we propose several algorithms to support capacity analysis and guide operator’s capacity planning tasks. Our algorithms are evaluated with real systems and experimental results are also included to demonstrate the effectiveness of our approach.
Similar content being viewed by others
References
Almeida, V., Menasce, D.: Capacity planning: An essential tool for managing web services. IEEE IT Prof. 4(4), 33–38 (2002)
Amazon: http://phx.corporate-ir.net/phoenix.zhtml?c=97664&p=irol-newsArticle&ID=798960&highlight=
Barford, P., Crovella, M.: Generating representative web workloads for network and server performance evaluation. In: SIGMETRICS ’98/PERFORMANCE ’98: Proceedings of the 1998 ACM SIGMETRICS Joint International Conference on Measurement and Modeling of Computer Systems, pp. 151–160, 1998
Bracewell, R.: The Fourier Transform and Its Applications, 3nd edn. McGraw-Hill Science/Engineering/Math, New York (1999)
Brockwell, P., Davis, R.: Introduction to Time Series and Forecasting, 2nd edn. Springer, Berlin (2003)
Cohen, I., Goldszmidt, M., Kelly, T., Symons, J., Chase, J.: Correlating instrumentation data to system states: a building block for automated diagnosis and control. In: OSDI’04: Proceedings of the 6th Conference on Symposium on Operating Systems Design and Implementation, p. 16, 2004
Cormen, T., Leiserson, C., Rivest, R.: Introduction to Algorithms, 1st ed. MIT Press/McGraw-Hill, Cumberland, New York (1990)
Das, R., Kephart, J., Whalley, I., Vytas, P.: Towards commercialization of utility-based resource allocation. In: The 3rd International Conference on Autonomic Computing (ICAC2006), pp. 287–290, Dublin, Ireland, June 2006
Fan, X., Weber, W.-D., Barroso, L.A.: Power provisioning for a warehouse-sized computer. In: ISCA ’07: Proceedings of the 34th Annual International Symposium on Computer Architecture, pp. 13–23, San Diego, California, USA, 2007
Hellerstein, J., Diao, Y., Parekh, S., Tilbury, D.M.: Feedback Control of Computing Systems. Wiley-IEEE Press, New York (2004)
IBM. http://www-935.ibm.com/services/us/its/pdf/g563-0339-00.pdf
JBoss. http://www.jboss.org
Jiang, G., Chen, H., Yoshihira, K.: Discovering likely invariants of distributed transaction systems for autonomic system management. In: The 3rd International Conference on Autonomic Computing (ICAC2006), pp. 199–208, Dublin, Ireland, June 2006
Jiang, G., Chen, H., Yoshihira, K.: Modeling and tracking of transaction flow dynamics for fault detection in complex systems. IEEE Trans. Dependable Secure Comput. 3(4), 312–326 (2006)
Jiang, G., Chen, H., Yoshihira, K.: Efficient and scalable algorithms for inferring likely invariants in distributed systems. IEEE Trans. Knowl. Data Eng. 19(11) (2007)
Kant, K., Won, Y.: Server capacity planning for web traffic workload. IEEE Trans. Knowl. Data Eng. 11(5), 731–747 (1999)
Kephart, J., Chess, D.: The vision of autonomic computing. Computer 36(1), 41–52 (2003)
Kuo, B.: Automatic Control Systems, 6th edn. Prentice-Hall, Englewood (1991)
Kusic, D., Kandasamy, N.: Risk-aware limited lookahead control for dynamic resource provisioning in enterprise computing systems. In: The 3rd International Conference on Autonomic Computing (ICAC2006), pp. 74–83, Dublin, Ireland, June 2006
Ljung, L.: System Identification—Theory for The User, 2nd edn. Prentice Hall PTR, New York (1998)
Menasce, D., Dowdy, L., Almeida, V.: Performance by Design: Computer Capacity Planning By Example, 1st ed. Prentice Hall PTR, New York (2004)
Microsoft. http://office.microsoft.com/en-us/assistance/HA011647631033.aspx
Microsoft. http://technet.microsoft.com/en-us/library/aa997558.aspx
Microsoft Office 2003 system requirments. http://support.microsoft.com/kb/822129
Parekh, J., Jung, G., Swint, G., Pu, C., Sahai, A.: Comparison of performance analysis approaches for bottleneck detection in multi-tier enterprise applications. In: IEEE International Workshop on Quality of Services, pp. 302–306, New Haven, CT, USA, 2006
Rissanen, J.: Stochastic Complexity in Statistical Inquiry Theory. World Scientific, Singapore (1989)
Stewart, C., Kelly, T., Zhang, A.: Exploiting nonstationarity for performance prediction. SIGOPS Oper. Syst. Rev. 41(3), 31–44 (2007)
Stewart, C., Shen, K.: Performance modeling and system management for multi-component online services. In: NSDI’05: Proceedings of the 2nd conference on Symposium on Networked Systems Design and Implementation, pp. 71–84, Boston, Massachusetts, USA, 2005
Urgaonkar, B., Pacifici, G., Shenoy, P., Spreitzer, M., Tantawi, A.: An analytical model for multi-tier internet services and its applications. SIGMETRICS Perform. Eval. Rev. 33(1), 291–302 (2005)
Urgaonkar, B., Pacifici, G., Shenoy, P., Spreitzer, M., Tantawi, A.: Analytic modeling of multitier internet applications. ACM Trans. Web 1(1), 2 (2007)
Walsh, W., Tesauro, G., Kephart, J., Das, R.: Utility functions in autonomic systems. In: The First International Conference on Autonomic Computing (ICAC2004), pp. 70–77, New York, May 2004
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Jiang, G., Chen, H. & Yoshihira, K. Profiling services for resource optimization and capacity planning in distributed systems. Cluster Comput 11, 313–329 (2008). https://doi.org/10.1007/s10586-008-0063-x
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10586-008-0063-x