The internet vs e-commerce servers: when will server performance matter?

ABSTRACT

The cycle time of an Internet based online shopper includes time at an electronic commerce (e-commerce) server to gather information and purchase products, download time to transfer data over the Internet, and think time for interpreting the results of individual requests. Currently most home based shoppers are limited to 56.6K modems and have cycle times largely determined by download time. Mega-bit (Mb) modems will soon be commonplace and will cause a significant reduction in the download time component of the shopper cycle time. This gain in download time can be utilized by the shoppers to submit additional requests to the server during their cycle times leading to an increased overall load on the server. The purpose of this paper is to consider the impact of higher shopper bandwidths on the performance of web-based shopping servers. To start with, we study the contents of several professionally managed e-commerce sites to obtain measures that include average page size and measures of mall size. The performance of a demonstration shopping mall system is measured under a controlled load to obtain optimistic measures of resource demands for such sites. An analytic model is developed and validated with respect to the measured system. The model is then modified to predict the behavior of both small and large shopping mall sites as client access to bandwidth and Internet performance increases.

REFERENCES

• 1: John Dilley , Rich Friedrich , Tai Jin , Jerome A. Rolia, Measurement Tools and Modeling Techniques for Evaluating Web Server Performance, Proceedings of the 9th International Conference on Computer Performance Evaluation: Modelling Techniques and Tools, p.155-168, June 03-06, 1997
• 2: {2} D. Krishnamurthy, Performance Characterization of Web-Based Shopping Systems, "MEng. Thesis", Department of Systems and Computer Engineering, Carleton University, Canada, September 1998.
• 3: {3} D. Krishnamurthy, J. Rolia, Workload Characterization Tools for E-Commerce Servers, In "Electronic Commerce: International IFIP/GI Working Conference on Trends in Distributed Systems for Electronic Commerce", pages 5-15, dpunkt.verlag, Hamburg, 1998.
• 4: Henrik Frystyk Nielsen , James Gettys , Anselm Baird-Smith , Eric Prud'hommeaux , Håkon Wium Lie , Chris Lilley, Network performance effects of HTTP/1.1, CSS1, and PNG, Proceedings of the ACM SIGCOMM '97 conference on Applications, technologies, architectures, and protocols for computer communication, p.155-166, September 14-18, 1997, Cannes, France
• 5: {5} D. Krishnamurthy and J. Rolia, Predicting the Performance of an E-Commerce Server: Those Mean Percentiles, Presented at the "Workshop on Internet Server Performance", Wisconsin, USA, June 1998.
• 6: C. Murray Woodside , John E. Neilson , Dorina C. Petriu , Shikharesh Majumdar, The Stochastic Rendezvous Network Model for Performance of Synchronous Client-Server-like Distributed Software, IEEE Transactions on Computers, v.44 n.1, p.20-34, January 1995 [doi>10.1109/12.368012]
• 7: J. A. Rolia , K. C. Sevcik, The Method of Layers, IEEE Transactions on Software Engineering, v.21 n.8, p.689-700, August 1995 [doi>10.1109/32.403785]
• 8: Edward D. Lazowska , John Zahorjan , G. Scott Graham , Kenneth C. Sevcik, Quantitative system performance: computer system analysis using queueing network models, Prentice-Hall, Inc., Upper Saddle River, NJ, 1984
• 9: {9} M. Reiser, A Queuing Network Analysis of Computer Communication Networks with Window Flow Control, "IEEE Transactions on Communications", pages 1201-1209, August 1979.
• 10: {10} J. Rolia, Predicting the Performance of Software Systems, "CSRI Technical Report 260", University of Toronto, Canada, January 1992.
• 11: {11} R. Jain, The Art of Computer Systems Performance Analysis, John Wiley and Sons, 1991.

INDEX TERMS

Primary Classification:
  C. Computer Systems Organization
  C.4 PERFORMANCE OF SYSTEMS
      Subjects: Modeling techniques

Additional Classification:
  C. Computer Systems Organization
  C.4 PERFORMANCE OF SYSTEMS
      Subjects: Performance attributes

  H. Information Systems
  H.3 INFORMATION STORAGE AND RETRIEVAL
      H.3.5 On-line Information Services
          Subjects: Web-based services; Commercial services

  K. Computing Milieux
  K.4 COMPUTERS AND SOCIETY
      K.4.4 Electronic Commerce
          Subjects: Electronic data interchange (EDI)

General Terms:
Algorithms, Design, Experimentation, Measurement, Performance, Theory