David Olshefski
Jason Nieh
Abstract
As businesses continue to grow their World Wide Web presence, it is becoming increasingly vital for them to have quantitative measures of the mean client perceived response times of their web services. We present Certes (CliEnt Response Time Estimated by the Server), an online server-based mechanism that allows web servers to estimate mean client perceived response time, as if measured at the client. Certes is based on a model of TCP that quantifies the effect that connection drops have on mean client perceived response time by using three simple server-side measurements: connection drop rate, connection accept rate and connection completion rate. The mechanism does not require modifications to HTTP servers or web pages, does not rely on probing or third party sampling, and does not require client-side modifications or scripting. Certes can be used to estimate response times for any web content, not just HTML. We have implemented Certes and compared its response time estimates with those obtained with detailed client instrumentation. Our results demonstrate that Certes provides accurate server-based estimates of mean client response times in HTTP 1.0/1.1 environments, even with rapidly changing workloads. Certes runs online in constant time with very low overhead. It can be used at websites and server farms to verify compliance with service level objectives.
- Allman, M. 2000. A web server's view of the transport layer. ACM Comput. Commun. Rev. 30, 4 (Oct.), 133--142.]] Google Scholar
- Almeida, J., Dabu, M., Manikutty, A., and Cao, P. 1998. Providing differentiated levels of service in web content hosting. In Tech. Rep. CS-TR-1998-1364. Computer Sciences Department, University of Wisconsin-Madison.]]Google Scholar
- Balakrishnan, H., Rahul, H. S., and Seshan, S. 1999. An integrated congestion management architecture for internet hosts. ACM SIGCOMM Comput. Commun. Rev. 29, 4 (Oct.), 175--187.]] Google Scholar
- Barford, P. and Crovella, M. 1999. A performance evaluation of hyper text transfer protocols. ACM SIGMETRICS Perf. Eval. Rev. 27, 1 (June), 188--197.]] Google Scholar
- Bhatti, N. and Friedrich, R. 1999. Web server support for tiered services. IEEE Net. 13, 5 (Sept.-Oct.), 64--71.]]Google Scholar
- Braden, R. 1989. RFC 1122: Requirements for Internet Hosts---Communication layers. IETF, http://www.ietf.org.]] Google Scholar
- Cardwell, N., Savage, S., and Anderson, T. 2000. Modeling TCP Latency. In IEEE INFOCOMM Conference Proceedings (Tel-Aviv, Israel). IEEE Computer Society Press, Los Alamitos, Calif., 1742--1751.]]Google Scholar
- Chen, X. and Mohapatra, P. 1999. Providing differentiated service from an internet server. In 8th International Conference on Computer Communications and Networks Conference Proceedings (Boston, Mass.). IEEE Computer Society Press, Los Alamitos, Calif., 214--217.]]Google Scholar
- Chen, X., Mohapatra, P., and Chen, H. 2001. An admission control scheme for predictable server response time for web accesses. In 10th International World Wide Web Conference Proceedings (Hong Kong, China). 545--554.]] Google Scholar
- Cohen, E., Krishnamurthy, B., and Rexford, J. 1999. Efficient algorithms for predicting requests to web servers. In IEEE INFOCOM Conference Proceedings (Orlando, Fla.). IEEE Computer Society Press, Los Alamitos, Calif., 284--293.]]Google Scholar
- Danzig, P. 2001. Ideas for next generation content delivery. In NOSSDAV 2001 (Port Jefferson, N.Y.). ACM, New York, http://www.nossdav.org/2001/keynote_nossdav2001.ppt.]]Google Scholar
- Eggert, L. and Heidemann, J. 1999. Application-level differentiated services for web servers. WWW J. 3, 2 (Aug.), 133--142.]] Google Scholar
- Exodus. http://www.exodus.com/.]]Google Scholar
- FreeBSD. http://www.FreeBSD.org/.]]Google Scholar
- Fu, Y., Cherkasova, L., Tang, W., and Vahdat, A. 2002. EtE: Passive end-to-end internet service performance monitoring. In USENIX Conference Proceedings (Monterey, Calif.). 115--130.]] Google Scholar
- Golub, G. H. and Loan, C. F. V. 1996. Matrix Computations. The John Hopkins University Press, Baltimore, Md.]] Google Scholar
- Kanodia, V. and Knightly, E. 2000. Multi-class latency-bounded web services. In IEEE/IFIP IWQoS Conference Proceedings (Pittsburgh, Pa).]]Google Scholar
- KeyNote. http://www.keynote.com/.]]Google Scholar
- Li, K. and Jamin, S. 2002. A measurement-based admission-controlled web server. In IEEE INFOCOMM Conference Proceedings IEEE, New York, NY, 651--659.]]Google Scholar
- Lu, C., Abdelzaher, T., Stankovic, J., and Son, S. H. 2001. A feedback control approach for guaranteeing relative delays in web server. In Proceedings of the 7th IEEE Real-Time Technology and Applications Symposium (Taipei, Taiwan). IEEE Computer Society Press, Los Alamitos, Calif.]] Google Scholar
- Mercury Interactive. http://www-heva.mercuryinteractive.com/.]]Google Scholar
- Microsoft. http://www.MicroSoft.com/.]]Google Scholar
- Mockapetris, P. 1987a. RFC 1034: Domain names concepts and facilities. IETF, http://www.ietf.org.]] Google Scholar
- Mockapetris, P. 1987b. RFC 1035: Domain names implementation and specification. IETF, http://www.ietf.org.]] Google Scholar
- Nahum, E., Barzilai, T., and Kandlur, D. 1999. Performance issues in WWW servers. ACM SIGMETRICS Performance Evaluation Review 27, 1 (May), 216--217.]] Google Scholar
- NetBSD. http://www.NetBSD.org/.]]Google Scholar
- NetQoS. http://www.NetQoS.com/.]]Google Scholar
- Nielsen, H. F., Gettys, J., Baird-Smith, A., Prud'hommeaux, E., Lie, H. W., and Lilley, C. 1997. Network performance effects of HTTP/1.1, CSS1, and PNG. ACM SIGCOMM Comput. Commun. Rev. 27, 4 (Oct.), 155--166.]] Google Scholar
- Olshefski, D., Nieh, J., and Agrawal, D. 2002. Inferring client response time at the web server. In ACM SIGMETRICS Conference Proceedings (Marina Del Rey, Calif.). ACM, New York, 160--171.]] Google Scholar
- OneStat. 2002. Microsoft's windows OS global market share is more than 97% according to OneStat.com. OneStat Press Release.]]Google Scholar
- Padhye, J., Firoiu, V., Towsley, D., and Kurose, J. 1998. Modeling TCP throughput: A simple model and its empirical validation. ACM SIGCOMM Comput. Commun. Rev. 28, 4 (Oct.), 303--314.]] Google Scholar
- Pahdye, J. and Floyd, S. 2001. On inferring TCP behavior. In Proceedings of the 2001 Conference on Applications, Technologies, Architectures, and Protocols for Computer Communications (San Diego, Calif.). ACM New York, 287--298.]] Google Scholar
- Pandey, R., Barnes, J. F., and Olsson, R. 1998. Supporting quality of service in HTTP servers. In Proceedings of the 17th annual ACM Symposium on Principles of Distributed Computing (Puerto Vallarta, Mexico). ACM New York, 247--256.]] Google Scholar
- Papoulis, A. and Pillai, S. U. 2001. Probability, Random Variables, and Stochastic Processes. McGraw-Hill Series in Electrical Engineering.]]Google Scholar
- Parekh, S., Gandhi, N., Hellerstein, J., Tilbury, D., Jayram, T., and Bigus, J. 2001. Using control theory to achieve service level objectives in performance management. In IFIP/IEEE International Symposium on Integrated Network Management Conference Proceedings (Seattle, Wash.). IEEE Computer Society Press, Los Alamitos, Calif. 841--854.]]Google Scholar
- Postel, J. 1981. RFC 793: Transmission Control Protocol. IETF, http://www.ietf.org.]]Google Scholar
- Press, W. H., Teukolsky, S. A., Vetterling, W. T., and Flannery, B. P. 1992. Numerical Recipes in C: The Art of Scientific Computing, 2nd Edition. Cambridge University Press, Cambridge, United Kingdom.]] Google Scholar
- Rajamony, R. and Elnozahy, M. 2001. Measuring client-perceived response times on the WWW. In 3rd USENIX Symposium on Internet Technologies and Systems (USITS) Conference Proceedings (USITS '01) (San Francisco, Calif.).]] Google Scholar
- RedHat. http://www.RedHat.com/.]]Google Scholar
- Rizzo, L. 1997. Dummynet: A simple approach to the evaluation of network protocols. ACM SIGCOMM Comput. Commun. Rev. 27, 1 (Jan.), 31--41.]] Google Scholar
- Smith, F. D., Campos, F. H., Jeffay, K., and Ott, D. 2001. What TCP/IP protocol headers can tell us about the web. ACM SIGMETRICS Perf. Eval. Rev. 29, 1 (June), 245--256.]] Google Scholar
- Srisuresh, P. and Egevang, K. 2001. RFC 3022: Traditional IP Network Address Translator (Traditional NAT). IETF, http://www.ietf.org.]] Google Scholar
- Srisuresh, P. and Holdredge, M. 1999. RFC 2663: IP Network Address Translator (NAT) Terminology and Considerations. IETF, http://www.ietf.org.]] Google Scholar
- Stevens, W. R. 1994. TCP/IP Illustrated, Volume 1 The Protocols. Addison-Wesley, Reading, Mass.]] Google Scholar
- StreamCheck. http://www.streamcheck.com/.]]Google Scholar
- Voigt, T., Tewari, R., Mehra, A., and Freimuth, D. 2001. Kernel mechanisms for service differentiation in overloaded web servers. In USENIX Conference Proceedings (Boston, Mass.). 189--202.]] Google Scholar
- WebStone. http://www.mindcraft.com/.]]Google Scholar
- Yajnik, M., Moon, S., Kurose, J., and Towsley, D. 1999. Measurement and modeling of the temporal dependence in packet loss. In IEEE INFOCOM Conference Proceedings (Orlando, Fla.). IEEE Computer Society Press, Los Alamitos, Calif., 345--352.]]Google Scholar
- Zhang, Y., Paxson, V., and Shenker, S. 2000. The Stationarity of Internet Path Properties: Routing, Loss and Throughput. In Tech. Rep. ACIRI.]]Google Scholar
Index Terms
- Using certes to infer client response time at the web server
Recommendations
Inferring client response time at the web server
SIGMETRICS '02: Proceedings of the 2002 ACM SIGMETRICS international conference on Measurement and modeling of computer systemsAs businesses continue to grow their World Wide Web presence, it is becoming increasingly vital for them to have quantitative measures of the client perceived response times of their web services. We present Certes (CliEnt Response Time Estimated by the ...
Inferring client response time at the web server
Measurement and modeling of computer systemsAs businesses continue to grow their World Wide Web presence, it is becoming increasingly vital for them to have quantitative measures of the client perceived response times of their web services. We present Certes (CliEnt Response Time Estimated by the ...
Understanding the management of client perceived response time
SIGMETRICS '06/Performance '06: Proceedings of the joint international conference on Measurement and modeling of computer systemsUnderstanding and managing the response time of web services is of key importance as dependence on the World Wide Web continues to grow. We present Remote Latency-based Management (RLM), a novel server-side approach for managing pageview response times ...
Comments