This Article 
 Bibliographic References 
 Add to: 
Server Capacity Planning for Web Traffic Workload
September/October 1999 (vol. 11 no. 5)
pp. 731-747

Abstract—The goal of this paper is to provide a methodology for determining bandwidth requirements for various hardware components of a World Wide Web server. The paper assumes a traditional symmetric multiprocessor (SMP) architecture for the web-server, although the same analysis applies to an SMP node in a cluster. The paper derives formulae for bandwidth demands for memory, processor data bus, network adapters, disk adapters, I/O-memory paths, and I/O buses. Since the web workload characteristics vary widely, three sample workloads are considered for illustrative purposes: 1) standard SPECweb96, 2) a SPECweb96-like workload that assumes dynamic data and retransmissions, and 3) WebProxy, which models a web proxy server that does not do much caching and, thus, has rather severe requirements. The results point to a few general conclusions regarding Web workloads. In particular, reduction in memory/data bus bandwidth by using the virtual interface architecture (VIA) is very desirable, and the connectivity needs may go well beyond the capabilities of traditional systems based on the traditional PCI-bus. Web workloads also demand a significantly higher memory bandwidth than data bus bandwidth and this disparity is expected to increase with the use of VIA. Also, the current efforts to offload TCP/IP processing may require a larger headroom in I/O subsystem bandwidth than in the processor-memory subsystem.

[1] M. Arlitt, R. Friedrich, and T. Jin, “Performance Evaluation of Web Proxy Cache Replacement Policies,” technical report, Hewlett Packard Laboratories, 1998.
[2] G. Bhal, “SPECweb96—A Benchmark for Web Server Performance,” Sun Microsystems, Oct. 1996.
[3] R. Caceres et al., “Web Proxy Caching: The Devil Is in the Details,” Proc. Usenix Symp. Internetworking Technologies and Systems, 1998.
[4] K. Claffy, G. Miller, and K. Thompson, “The Nature of the Beast: Recent Traffic Measurements from an Internet Backbone,” Cooperative Assoc. for Internet Data Analysis (CAIDA);
[5] D. Clark et al., "An Analysis of TCP Processing Overhead," IEEE Comm., vol. 27, no. 6, June 1989, pp. 23-29.
[6] M.E. Crovella and A. Bestavros, "Explaining World Wide Web Traffic Self-Similarity," Boston Univ. Tech. Report TR-95-015, 1995.
[7] B.M. Duska, D. Marwood, and M.J. Feeley, “The Measured Access Characteristics of World-Wide Web Client Proxy Caches,” Proc. Usenix Symp Internetworking Technologies and Systems, 1997.
[8] F. Douglis, A. Feldmann, and B. Krishnamurthy, “Rate of Change and Other Metrics: A Live Study of World Wide Web,” Proc. Usenix Symp. Internetworking Technologies and Systems, 1997.
[9] D. Dunning et al., "The Virtual Interface Architecture," IEEE Micro, Mar.-Apr. 1998, pp. 66-76.
[10] R. Felding et al., “Hypertext Transfer Protocol—HTTP 1. 1,” IETF RFC 2068, Jan. 1997.
[11] A. Feldmann et al., "Performance of Web Proxy Caching in Heterogeneous Bandwidth Environments," Proc. Joint Conf. IEEE Computer and Comm. Societies (Infocom), IEEE CS Press, 1999, pp. 107–116.
[12] J. Hennessy and D. Patterson, Computer Architecture: A Quantitative Approach. Morgan Kaufmann, 1995.
[13] R.W. Horst and D. Garcia, “ServerNet SAN I/O Architecture,” Proc. Hot Interconnects, 1997.
[14] K. Kant and Y. Won, “Server Capacity Planning for Web Traffic Workload,” technical report, Server Architecture Laboratory, Intel Corp., Sept. 1998,
[15] J. Mogul et al., "Potential Benefits of Delta-Encoding and Data Compression for HTTP," Proc. ACM SIGCOMM, ACM Press, New York, 1997, pp. 181-194. An extended and corrected version appears as Research Report 97/4a, Digital Equipment Corp. Western Research Laboratory, Dec. 1997.
[16] I. Norros, “A Buffer with Self-Similar Input,” Queuing Systems, vol. 16, no. 2, pp. 382-396, Feb. 1994.
[17] K. Park, G. Kim, and M. Crovella, “On the Effect of Traffic Self-Similarity on Network Performance,” Proc. SPIE Int'l Conf. Performance and Control of Network Systems, Nov. 1997.
[18] V. Paxson and S. Floyd, “Wide Area Traffic: The Failure of Poisson Modeling,” IEEE/ACM Trans. Networking, vol. 3, no. 3, pp. 226-244, June 1995.
[19] W. Leland et al., "On the Self-Similar Nature of Ethernet Traffic (Extended Version)," IEEE/ACM Trans. Networking, Vol. 2, No. 1, Feb. 1994, pp. 1-15.
[20] W. Willinger, M.S. Taqqu, R. Sherman, and D.V. Wilson, “Self-Similarity through High Variability: Statistical Analysis of Ethernet LAN Traffic at the Source Level,” Proc. SIGCOMM '95, pp. 100-113, 1995.

Index Terms:
Web server, traffic characterization, self-similarity, symmetric multiprocessors, caching/proxy server, band-width requirements.
Krishna Kant, Youjip Won, "Server Capacity Planning for Web Traffic Workload," IEEE Transactions on Knowledge and Data Engineering, vol. 11, no. 5, pp. 731-747, Sept.-Oct. 1999, doi:10.1109/69.806933
Usage of this product signifies your acceptance of the Terms of Use.