Measuring and characterizing end-to-end Internet service performance

ABSTRACT

Fundamental to the design of reliable, high-performance network services is an understanding of the performance characteristics of the service as perceived by the client population as a whole. Understanding and measuring such end-to-end service performance is a challenging task. Current techniques include periodic sampling of service characteristics from strategic locations in the network and instrumenting Web pages with code that reports client-perceived latency back to a performance server. Limitations to these approaches include potentially nonrepresentative access patterns in the first case and determining the location of a performance bottleneck in the second.This paper presents EtE monitor, a novel approach to measuring Web site performance. Our system passively collects packet traces from a server site to determine service performance characteristics. We introduce a two-pass heuristic and a statistical filtering mechanism to accurately reconstruct different client page accesses and to measure performance characteristics integrated across all client accesses. Relative to existing approaches, EtE monitor offers the following benefits: i) a latency breakdown between the network and server overhead of retrieving a Web page, ii) longitudinal information for all client accesses, not just the subset probed by a third party, iii) characteristics of accesses that are aborted by clients, iv) an understanding of the performance breakdown of accesses to dynamic, multitiered services, and v) quantification of the benefits of network and browser caches on server performance. Our initial implementation and performance analysis across three different commercial Web sites confirm the utility of our approach.

REFERENCES

• 1: Paul Barford , Mark Crovella, Critical path analysis of TCP transactions, Proceedings of the conference on Applications, Technologies, Architectures, and Protocols for Computer Communication, p.127-138, August 28-September 01, 2000, Stockholm, Sweden
• 2: Caceres, R., Duffield, N., Feldmann, A., Friedmann, J., Greenberg, A., Greer, R., Johnson, T., Kalmanek, C., Krishnamurthy, B., Lavelle, D., Mishra, P., Ramakrishnan, K., Rexford, J., True, F., and van der Merwe, J. 2000. Measurement and Analysis of IP Network Usage and Behaviour.
• 3: Candle Corporation: eBusiness Assurance. http://www.candle.com/.
• 4: Cardwell, N., Savage, S., and Anderson, T. 2000. Modeling TCP Latency. In INFOCOM.
• 5: Cisco Distributed Director. http://www.cisco.com/.
• 6: Anja Feldmann, BLT: Bi-layer tracing of HTTP and TCP&slash;IP, Proceedings of the 9th international World Wide Web conference on Computer networks : the international journal of computer and telecommunications netowrking, p.321-335, June 2000, Amsterdam, The Netherlands
• 7: Fielding, R., Gettys, J., Mogul, J., Nielsen, H., and Berners-Lee, T. 2001. Hypertext Transfer Protocol---HTTP/1.1. Tech. Rep. RFC 2616, IETF. June.
• 8: Gomez, Inc. http://www.gomez.com.
• 9: ETE: A Customizable Approach to Measuring End-to-End Response Times and Their Components in Distributed Systems, Proceedings of the 19th IEEE International Conference on Distributed Computing Systems, p.152, May 31-June 04, 1999
• 10: HP Corporation. OpenView Products: Web Transaction Observer. http://www.openview.hp.com.
• 11: IBM Corporation. Tivoli Web Management Solutions. http://www.tivoli.com/products/demos/twsm.html.
• 12: IBM Research. Page Detailer. http://www.research.ibm.com/pagedetailer/.
• 13: JavaServer Pages. http://java.sun.com/products/jsp/technical.html.
• 14: JavaServlet Technology. http://java.sun.com/products/servlet/.
• 15: Terry Keeley, Thin, High Performance Computing over the Internet, Proceedings of the 8th International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems, p.407, August 29-September 01, 2000
• 16: Keynote Systems, Inc. http://www.keynote.com.
• 17: Krishnamurthy, B. and Rexford, J. 1998. Software Issues in Characterizing Web Server Logs.
• 18: Krishnamurthy, B. and Rexford, J. 1999. En Passant: Predicting HTTP/1.1 Traffic.
• 19: Web protocols and practice: HTTP/1.1, Networking protocols, caching, and traffic measurement, Addison-Wesley Longman Publishing Co., Inc., Boston, MA, 2001
• 20: Balachander Krishnamurthy , Jia Wang, On network-aware clustering of Web clients, Proceedings of the conference on Applications, Technologies, Architectures, and Protocols for Computer Communication, p.97-110, August 28-September 01, 2000, Stockholm, Sweden
• 21: Balachander Krishnamurthy , Yin Zhang , Craig E. Wills , Kashi Vishwanath, Design, implementation, and evaluation of a client characterization driven web server, Proceedings of the 12th international conference on World Wide Web, May 20-24, 2003, Budapest, Hungary [doi>10.1145/775152.775172]
• 22: Balachander Krishnamurthy , Craig E. Wills, Analyzing factors that influence end-to-end Web performance, Proceedings of the 9th international World Wide Web conference on Computer networks : the international journal of computer and telecommunications netowrking, p.17-32, June 2000, Amsterdam, The Netherlands
• 23: Balachander Krishnamurthy , Craig E. Wills, Improving web performance by client characterization driven server adaptation, Proceedings of the 11th international conference on World Wide Web, May 07-11, 2002, Honolulu, Hawaii, USA [doi>10.1145/511446.511486]
• 24: David Mosberger , Tai Jin, httperf—a tool for measuring web server performance, ACM SIGMETRICS Performance Evaluation Review, v.26 n.3, p.31-37, Dec. 1998 [doi>10.1145/306225.306235]
• 25: NetMechanic, Inc. http://www.netmechanics.com.
• 26: NetQoS Inc. http://www.netqos.com.
• 27: 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
• 28: Olshefski, D. P., Nieh, J., and Agrawal, D. 2001. Inferring Client Response Time at the Web Server. In Proceedings of USITS.
• 29: Porivo Technologies, Inc. http://www.porivo.com.
• 30: Rajamony, R. and Elnozahy, M. 2001. Measuring Client-Perceived Response Times on the WWW. In Proceedings of the Third USENIX Symposium on Internet Technologies and Systems (USITS).
• 31: Seshan, S., Stemm, M., and Katz, R. H. 1997. SPAND: Shared Passive Network Performance Discovery. In Proceedings of USITS.
• 32: F. Donelson Smith , Félix Hernández Campos , Kevin Jeffay , David Ott, What TCP/IP protocol headers can tell us about the web, Proceedings of the 2001 ACM SIGMETRICS international conference on Measurement and modeling of computer systems, p.245-256, June 2001, Cambridge, Massachusetts, United States
• 33: Software Research Inc. http://www.soft.com.
• 34: Stemm, M., Katz, R. H., and Seshan, S. 2000. A Network Measurement Architecture for Adaptive Applications. In Proceedings of IEEE INFOCOM.
• 35: Tcpdump. http://www.tcpdump.org.

INDEX TERMS

Primary Classification:
  C. Computer Systems Organization
  C.2 COMPUTER-COMMUNICATION NETWORKS
      C.2.3 Network Operations
          Subjects: Network monitoring

Additional Classification:
  C. Computer Systems Organization
  C.2 COMPUTER-COMMUNICATION NETWORKS
      C.2.4 Distributed Systems
          Subjects: Client/server
      C.2.5 Local and Wide-Area Networks
          Subjects: Internet (e.g., TCP/IP)
  C.4 PERFORMANCE OF SYSTEMS
      Subjects: Design studies; Measurement techniques; Modeling techniques

  D. Software
  D.2 SOFTWARE ENGINEERING
      D.2.5 Testing and Debugging
          Subjects: Monitors
      D.2.8 Metrics
          Subjects: Performance measures

General Terms:
Measurement, Performance

Keywords:
End-to-end service performance, QoS, network packet traces, passive monitoring, reconstruction of web page composition, web site performance