Abstract
While many application service providers have proposed using thin-client computing to deliver computational services over the Internet, little work has been done to evaluate the effectiveness of thin-client computing in a wide-area network. To assess the potential of thin-client computing in the context of future commodity high-bandwidth Internet access, we have used a novel, noninvasive slow-motion benchmarking technique to evaluate the performance of several popular thin-client computing platforms in delivering computational services cross-country over Internet2. Our results show that using thin-client computing in a wide-area network environment can deliver acceptable performance over Internet2, even when client and server are located thousands of miles apart on opposite ends of the country. However, performance varies widely among thin-client platforms and not all platforms are suitable for this environment. While many thin-client systems are touted as being bandwidth efficient, we show that network latency is often the key factor in limiting wide-area thin-client performance. Furthermore, we show that the same techniques used to improve bandwidth efficiency often result in worse overall performance in wide-area networks. We characterize and analyze the different design choices in the various thin-client platforms and explain which of these choices should be selected for supporting wide-area computing services.
- 100×100 Project. The 100 × 100 Project. http://100x100network.org/.Google Scholar
- Allman, M., Floyd, S., and Partridge, C. 1998. RFC 2414: Increasing TCP's Initial Window. Google Scholar
- AT&T Laboratories Cambridge. Virtual Network Computing. http://www.uk.research.att.com/vnc.Google Scholar
- Ausbeck, P. 1999. A Streaming Piecewise-Constant Model. In Data Compression Conference (DCC), Snowbird, UT. Google Scholar
- Bradner, S. 2005. Implications of an improving Internet. Network World. http://www.networkworld.com/columnists/2005/040405bradner.html.Google Scholar
- broadwayinfo.com. Broadway / X Web FAQ. http://www.broadwayinfo.com/bwfaq.htm.Google Scholar
- Charon Systems. Charon Systems. http://www.charon.com.Google Scholar
- Christiansen, B. O., Schauser, K. E., and Munke, M. 2000. A Novel Codec for Thin Client Computing. In Data Compression Conference (DCC), Snowbird, UT. Google Scholar
- Citrix Systems 1998. Citrix MetaFrame 1.8 Backgrounder. Citrix White Paper, Citrix Systems.Google Scholar
- Corbato, F. J. and Vyssotsky, V. A. 1965. Introduction and Overview of the Multics System. In Proceedings of the Fall Joint Computer Conference. Vol. 27. 185--196. Google Scholar
- Cottrell, L. 2005. ICFA SCIC Network Monitoring Report. Tech. rep., International Committee for Future Accelerators (ICFA) and Standing Committee on Inter-Regional Connectivity (SCIC). Feb. http://www.slac.stanford.edu/xorg/icfa/icfa-net-paper-jan05/.Google Scholar
- Cumberland, B. C., Carius, G., and Muir, A. 1999. Microsoft Windows NT Server 4.0, Terminal Server Edition: Technical Reference. Microsoft Press, Redmond, WA.Google Scholar
- Danskin, J. and Hanrahan, P. 1994. Profiling the X Protocol. In Proceedings of the SIGMETRICS Conference on Measurement and Modeling of Computer Systems. Nashville, TN. Google Scholar
- Expertcity, Inc. 2000. DesktopStreaming Technology and Security. Expertcity White Paper.Google Scholar
- Howard, B. 2000. Thin Is Back. PC Magazine 19, 7 (July).Google Scholar
- J. G. Cleary, W. J. Teahan, I. H. W. 1995. Unbounded Length Contexts for PPM. Data Compression Conference (DCC), Snowbird, UT. Google Scholar
- Kaplinsk, C. Tight Encoding. http://www.tightvnc.com/compare.html.Google Scholar
- Lai, A. and Nieh, J. 2002. Limits of Wide-Area Thin-Client Computing. In Proceedings of the 2002 ACM SIGMETRICS International Conference on Measurement and Modeling of Computer Systems. ACM Press, Marina del Rey, CA, USA, 228--239. Google Scholar
- Lakshman, T. V. and Madhow, U. 1997. The Performance of TCP/IP for Networks with High Bandwidth-Delay Products and Random Loss. IEEE/ACM Trans. Netw. 5, 3 (June), 336-- 350. Google Scholar
- Lakshman, T. V., Madhow, U., and Suter, B. 2000. TCP/IP Performance with Random Loss and Bidirectional Congestion. IEEE/ACM Trans. Netw. 8, 5 (Oct.), 541--555. Google Scholar
- LapLink 1999. LapLink 2000 User's Guide. LapLink, Bothell, WA.Google Scholar
- Legard, D. 2003. Korea to build 100M bps Internet system. InfoWorld. http://www.infoworld.com/article/03/11/18/HNkorea_1.html.Google Scholar
- Mahdavi, J. Enabling High Performance Data Transfers on Hosts. http://www.psc.edu/networking/perf_tune.html.Google Scholar
- Mathers, T. W. and Genoway, S. P. 1998. Windows NT Thin Client Solutions: Implementing Terminal Server and Citrix MetaFrame. Macmillan Technical Publishing, Indianapolis, IN. Google Scholar
- Microsoft Corporation. Description of Windows 2000 TCP Features. http://support.microsoft.com/support/kb/articles/Q224/8/29.ASP.Google Scholar
- Microsoft Corporation 1998. Microsoft Windows NT Server 4.0, Terminal Server Edition: An Architectural Overview. Technical White Paper. Google Scholar
- National Laboratory for Applied Network Research a. Active data query. http://amp.nlanr.net/Active/raw_data/cgi-bin/data_form.cgi.Google Scholar
- National Laboratory for Applied Network Research b. Active measurements. http://amp.nlanr.net/.Google Scholar
- Nieh, J. and Yang, S. J. 2000. Measuring the Multimedia Performance of Server-Based Computing. In Proceedings of the 10th International Workshop on Network and Operating System Support for Digital Audio and Video. Chapel Hill, NC, 55--64.Google Scholar
- Nieh, J., Yang, S. J., and Novik, N. 2000. A Comparison of Thin-Client Computing Architectures. Tech. Rep. CUCS-022-00, Department of Computer Science, Columbia University. Nov.Google Scholar
- Nieh, J., Yang, S. J., and Novik, N. 2003. Measuring Thin-Client Performance Using Slow-Motion Benchmarking. ACM Trans. Comput. Sys. 21, 1 (Feb.), 87--115. Google Scholar
- Nielsen, J. 2000. Designing Web Usability: The Practice of Simplicity. New Riders Publishing, Indianapolis, Indiana. Google Scholar
- Richardson, T., Stafford-Fraser, Q., Wood, K. R., and Hopper, A. 1998. Virtual Network Computing. IEEE Internet Comput. 2, 1 (Jan./Feb.). Google Scholar
- Runaware.com. Runaware.com. http://www.runaware.com.Google Scholar
- Santa Cruz Operation 1998. Tarantella Web-Enabling Software: The Adaptive Internet Protocol. SCO Technical White Paper.Google Scholar
- Sayood, K. 2000. Introduction to Data Compression, 2nd Edition ed. Morgan Kaufmann Publishers, San Francisco, CA. Google Scholar
- Scheifler, R. W. and Gettys, J. 1986. The X Window System. ACM Trans. Graph. 5, 2 (Apr.), 79--106. Google Scholar
- Schmidt, B. K., Lam, M. S., and Northcutt, J. D. 1999. The Interactive Performance of SLIM: A Stateless, Thin-Client Architecture. In Proceedings of the 17th ACM Symposium on Operating Systems Principles ( SOSP ). Vol. 34. Kiawah Island Resort, SC, 32--47. Google Scholar
- Semke, J., Mahdavi, J., and Mathis., M. 1998. Automatic TCP Buffer Tuning. In Proceedings of ACM SIGCOMM '98. Vancouver, Canada, 315--323. Google Scholar
- Shaw, A., Burgess, K. R., Pullan, J. M., and Cartwright, P. C. 2000. Method of Displaying an Application on a Variety of Client Devices in a Client/Server Network. U.S. Patent US6104392.Google Scholar
- Shneiderman, B. 1992. Designing the User Interface: Strategies for Effective Human-Computer Interaction, 2nd ed. Addison-Wesley, Reading, MA. Google Scholar
- Shunra Software. The Cloud. http://www.shunra.com.Google Scholar
- Sun Microsystems. Sun Ray 1 Enterprise Appliance. http://www.sun.com/products/sunray1.Google Scholar
- Symantec Corporation. PC Anywhere. http://www.symantec.com/pcanywhere.Google Scholar
- The Trustees of Indiana University. Abilene Weather Map. http://hydra.uits.iu.edu/~abilene/traffic/.Google Scholar
- Tirumala, A. and Ferguson, J. Iperf. http://dast.nlanr.net/Projects/Iperf/.Google Scholar
- Tolly Research 2000. Thin-Client Networking: Bandwidth Consumption Using Citrix ICA. IT clarity. Google Scholar
- Weinberger, M., Seroussi, G., and Sapiro, G. 2000. The LOCO-I Lossless Image Compression Algorithm: Principles and Standardization into JPEG-LS. In IEEE Trans. Image Proc. Vol. 9. 1309--1324. Google Scholar
- WildPackets, Inc. Etherpeek 4. http://www.wildpackets.com.Google Scholar
- Wong, A. Y. and Seltzer, M. 1999. Evaluating Windows NT Terminal Server Performance. In Proceedings of the 3rd USENIX Windows NT Symposium. Seattle, WA, 145--154. Google Scholar
- Wong, A. Y. and Seltzer, M. 2000. Operating System Support for Multi-User, Remote, Graphical Interaction. In Proceedings of the USENIX 2000 Annual Technical Conference. San Diego, CA, 183--196. Google Scholar
- Yang, S. J. and Nieh, J. 2000. Thin Is In. PC Magazine 19, 13 (July), 68.Google Scholar
- Yang, S. J., Nieh, J., Krishnappa, S., Mohla, A., and Sajjadpour, M. 2003. Web Browsing Performance of Wireless Thin-Client Computing. In Proceedings of the Twelfth International World Wide Web Conference (WWW 2003). Budapest, Hungary. Google Scholar
- Yang, S. J., Nieh, J., Selsky, M., and Tiwari, N. 2002. The Performance of Remote Display Mechanisms for Thin-Client Computing. In Proceedings of the 2002 USENIX Annual Technical Conference. Monterey, CA, USA. Google Scholar
- Ziff-Davis, Inc. i-Bench version 1.5. http://etestinglabs.com/benchmarks/i-bench/i-bench.asp.Google Scholar
- Ziv, J. and Lempel, A. 1977. A Universal Algorithm for Data Compression. IEEE Trans. Inf. Theory 23, 3 (May), 337--343.Google Scholar
- Ziv, J. and Lempel, A. 1978. Compression of Individual Sequences via Variable-Rate Coding. IEEE Trans. Inf. Theory 24, 5 (Sept.), 530--536.Google Scholar
Index Terms
- On the performance of wide-area thin-client computing
Recommendations
Limits of wide-area thin-client computing
SIGMETRICS '02: Proceedings of the 2002 ACM SIGMETRICS international conference on Measurement and modeling of computer systemsWhile many application service providers have proposed using thin-client computing to deliver computational services over the Internet, little work has been done to evaluate the effectiveness of thin-client computing in a wide-area network. To assess ...
Measuring thin-client performance using slow-motion benchmarking
Modern thin-client systems are designed to provide the same graphical interfaces and applications available on traditional desktop computers while centralizing administration and allowing more efficient use of computing resources. Despite the rapidly ...
Limits of wide-area thin-client computing
Measurement and modeling of computer systemsWhile many application service providers have proposed using thin-client computing to deliver computational services over the Internet, little work has been done to evaluate the effectiveness of thin-client computing in a wide-area network. To assess ...
Comments