An optimization framework for web farm configuration

ABSTRACT

A common problem that sales consultants face in the field is the selection of an appropriate hardware and software configuration for web farms. Over-provisioning means that the tender will be expensive while under-provisioning will lead to a configuration that does not meet the customer criteria. Indy is a performance modeling environment which allows developers to create custom modeling applications. We have constructed an Indy-based application for defining web farm workloads and topologies. The paper presents an optimization framework that allows the consultant to easily find configurations that meet customers' criteria. The system searches the solution space creating possible configurations, using the web farm models to predict their performance. The optimization tool is then employed to select an optimal configuration. Rather than using a fixed algorithm, the framework provides an infrastructure for implementing multiple optimization algorithms. In this way, the appropriate algorithm can be selected to match the requirements of different types of problem. The framework incorporates a number of novel techniques, including caching results between problem runs, an XML based configuration language, and an effective method of comparing configurations. We have applied the system to a typical web farm configuration problem and results have been obtained for three standard optimization algorithms.

REFERENCES

• 1: Sigrún Andradóttir, A review of simulation optimization techniques, Proceedings of the 30th conference on Winter simulation, p.151-158, December 13-16, 1998, Washington, D.C., United States
• 2: Farhad Azadivar, Simulation optimization methodologies, Proceedings of the 31st conference on Winter simulation: Simulation---a bridge to the future, p.93-100, December 05-08, 1999, Phoenix, Arizona, United States [doi>10.1145/324138.324168]
• 3: J. Boesel, B.L. Nelson and N. Ishii, A Framework for Simulation-Optimization Software, Technical Report, Department of Industrial Engineering and Management Sciences, Northwestern University, 1999.
• 4: Justin Boesel , Royce O. Bowden, Jr. , Fred Glover , James P. Kelly , Erik Westwig, Panel: simulation optimization: future of simulation optimization, Proceedings of the 33nd conference on Winter simulation, December 09-12, 2001, Arlington, Virginia
• 5: A. Charnes and W.W. Cooper, Chance Constraint Programming, Management Science Vol 6, No 1, 1959.
• 6: E.K.P. Chong, R.L. Givan, H.S. Chang, A Framework for Simulation-based Network Control via Hindsight Optimization, Conference on Decision and Control (CDC), 2000.
• 7: M.C. Fu, Optimization via Simulation: A Review, Annals of Operations Research, 53:199-247.
• 8: Michael C. Fu, A tutorial review of techniques for simulation optimization, Proceedings of the 26th conference on Winter simulation, p.149-156, December 11-14, 1994, Orlando, Florida, United States
• 9: Björn Gehlsen , Bernd Page, A framework for distributed simulation optimization, Proceedings of the 33nd conference on Winter simulation, December 09-12, 2001, Arlington, Virginia
• 10: F. Glover, Tabu Search --- Part I, ORSA Journal of Computing, 190-206, Vol 1. No. 3, Summer 1989.
• 11: F. Glover, Tabu Search --- Part II, ORSA Journal of Computing, 4-32, Vol 2. No 1, Winter 1990.
• 12: J.C. Hardwick, E. Papaefstathiou, and D. Guimbellot, Modeling the Performance of E-Commerce Sites, Journal of Computer Resource Management, Winter 2002 Edition, Issue 105, pp. 3-12.
• 13: David G. Humphrey , James R. Wilson, A revised simplex search procedure for stochastic simulation response-surface optimization, Proceedings of the 30th conference on Winter simulation, p.751-760, December 13-16, 1998, Washington, D.C., United States
• 14: B.D. Joshi, R Unal, N.H.White and W.D. Morris, A Framework for the Optimization of Discrete-Event Simulation Models, Proceedings of the 17th ASEM National Conference, October 10-12 1996.
• 15: R. Jain, The Art of Computer Systems Performance Analysis (John Wiley & Sons, 1991).
• 16: H. Gonda Neddermeijer , Gerrit J. van Oortmarssen , Nanda Piersma , Rommert Dekker, A framework for Response Surface Methodology for simulation optimization, Proceedings of the 32nd conference on Winter simulation, December 10-13, 2000, Orlando, Florida
• 17: E. Papaefstathiou, Design of a performance technology infrastructure to support the construction of responsive software, Proceedings of the 2nd international workshop on Software and performance, p.96-104, September 2000, Ottawa, Ontario, Canada [doi>10.1145/350391.350414]
• 18: James R. Swisher , Paul D. Hyden , Sheldon H. Jacobson , Lee W. Schruben, Simulation optimization: a survey of simulation optimization techniques and procedures, Proceedings of the 32nd conference on Winter simulation, December 10-13, 2000, Orlando, Florida
• 19: W.T. Thompkins, Policy Driven, Multi-Objective Optimization of Simulated Stochastic Models, http://www.ncsa.uiuc.edu/TechFocus/Projects/NCSA/Policy Driven, Multi-Objective Optimization of Simulated Stochastic Models.html.

INDEX TERMS

Keywords:
design, experimentation, indy, infrastructures, measurement, modeling, optimization, performance, simulation