|
Vol: 53(67) No: 3 / September 2008 QBD Processes for Web Systems Raluca Popa Department of Automation and Applied Informatics, Budapest University of Technology and Economics, 1111 Budapest Goldmann György tér 3. IV. em., Budapest, Hungary, phone: (00361) 463-2870, e-mail: praluca@aut.bme.hu, web: http://www.aut.bme.hu/portal/praluca Márk Kaszó Department of Automation and Applied Informatics, Budapest University of Technology and Economics, 1111 Budapest Goldmann György tér 3. IV. em., Budapest, Hungary, e-mail: mkaszo@aut.bme.hu, web: http://www.aut.bme.hu/portal/mkaszo Keywords: Customer Behavior Model Graph (CBMG), Markov chain, Phase-type, MAP, QBD Abstract – Nowadays E-Market forms a crucial part of our life; there are a lot of E-Commerce sites available worldwide. The customers use different navigation patterns, thus they make different workload for E-Commerce systems. Performance measurements can be the base for performance modeling and prediction. With the help of performance models, the performance metrics can be determined at the early stages of the development process. Our aim is to characterize typical user workload to reproduce user behavior before the site is developed, even during a modeling process. In this paper we will present some mathematical processes to model web service interactions and a Performance Measure Engine which can be used easily to measure key performance metrics like CPU load or disk transfer. References [1] Latouche, G. and Taylor, P. editors. Matrix analytic Methods: Theory and Applications : Proceedings of the Fourth International Conference, Adelaide, 2002. [2] Menasce, D.A., Almeida, V. A. F. Scaling for E Business: Technologies, Models, Performance, and Capacity Planning. Prentice Hall, New York, 2000. [3] Jain R. K. The art of computer systems performance analysis. Wiley, New York, 1992. [4] Kaszó, M., Legány, C. Analyzing Customer Behavior Model Graph (CBMG) using Markov Chains 11th IEEE International Conference on Intelligent Engineering Systems, 71-76, 2007. [5] Popa, R., Kaszó, M. Performance and Scalability in ASP.NET Applications Automation and Applied Computer Science Workshop (AACS), 71-76, 2007. [6] Dudin, A.N., Kazimirsky, A.V., Klimenok, V.I. The Queueing Model MAP/PH/1/N with Feedback Operating in a Markovian Random Environment Austrian Journal of Statistics, 101-110, 2005. [7] Smith, W. D. Benchmark for E-Commerce Solution Internet Computing, IEEE, 2002. [8] Menasce, D.A. TPC-W: a benchmark for e-commerce Internet Computing, IEEE, 83-87, 2002. [9] Curbera, F. et al Unraveling the Web Services Web: An Introduction to SOAP, WSDL, and UDDI IEEE Internet Computing, 8693., 2002. [10] Riska, A., Smirni, E. M/G/1-type processes: A tutorial Performance Evaluation of Complex Computer Systems; Techniques and Tools, 36-63, 2002. [11] Riska, A., Squillante, M., Yu, S., Liu, Z., Zhang, L. Matrix-Analytic Analysis of a MAP/PH/1 Queue _tted to web server data Matrix-Analytic Methods: Theory and Applications, 335-356, 2002. [12] Xia, C. H., Liu, Z., Squillante, M. S., Zhang, L., Mabuch, N. Traffic Modeling and Performance Analysis of Commercial Web Sites Preprint, 2001. [13] Horvath, A., Rozsa, G., Telek, M. A MAP Fitting Method to Approximate Real Traffic Behavior Proceedings of IFIP Workshop on Performance Modelling and Evaluation of ATM IP networks, 32/1-12, 2000. [14] Telek, M., Horvath, A. Approximating heavy tailed behaviour with Phase type distributions Advances in Algorithmic Methods for Stochastic Methods, 32/1-12, 2000. |