IMPLEMENTATION OF A NEW SCHEDULING POLICY IN WEB SERVERS

Ahmad S. Al Sa'deh, Adnan H. Yahya

2008

Abstract

Recently, the Shortest-Remaining-Processing-Time (SRPT) has been proposed for scheduling static HTTP requests in web servers to reduce the mean response time. The SRPT assumes that the response time of the requested file is strongly proportional to its size. This assumption is unwarranted in Internet environment. Thus, we proposed the Shortest-Remaining-Response-Time (SRRT) that better estimates the response time for static HTTP. The SRRT prioritizes requests based on a combination of the current round-trip-time (RTT), TCP congestion window size (cwnd) and the size of what remains of the requested file. We compare SRRT to SRPT and Processor-Sharing (PS) policies. The SRRT shows the best improvement in the mean response time. SRRT gives an average improvement of about 7.5% over SRPT. This improvement comes at a negligible expense in response time for long requests. We found that under 100Mbps link, only 1.5% of long requests have longer response times than under PS. The longest request under SRRT has an increase in response time by a factor 1.7 over PS. For 10Mbps link, only 2.4% of requests are penalized, and SRRT increases the longest request time by a factor 2.2 over PS.

References

  1. King, A., 2003. Speed up your site: web site optimization, New Riders. Indiana, 1st edition.
  2. Nielsen, J., 1997. The need for speed. <http://www.useit.com/alertbox/9703a.html>.
  3. Kegel, D., 2006. The C10K problem. <http://www.kegel.com/c10k.html>.
  4. Crovella, M., Frangioso, R., 1999. Connection scheduling in web servers. In USENIX Symposium on Internet Technologies and Systems.
  5. Rawat, M., Kshemkayani, A., 2003. SWIFT: Scheduling in web servers for fast response time. In Second IEEE International Symposium on Network Computing and Applications.
  6. Jechlitschek, C., Gorinsky, S., 2007. Fair Efficiency, or Low Average Delay without Starvation. In Computer Communications and Networks, 2007.
  7. Bansal, N., Harchol-Balter, M., 2001. Analysis of SRPT scheduling: investigating unfairness. In ACM SIGMETRICS Performance Evaluation Review, vol. 29, no. 1, pp. 279-290.
  8. Manley, S., Seltzer, M., 1997. Web facts and fantasy. In Proceedings of the 1997 USITS-97.
  9. Murta, C., Corlassoli, T., 2003. Fastest connection first: A new scheduling policy for web servers. In The 18th International Teletra#c Congress (ITC-18).
  10. Schrage, L., Miller, L., 1966. The queue M/G/1 with the shortest remaining processing time discipline. In Operations Research, vol. 14, no. 4, pp. 670-684.
  11. Schrage, L., 1968. A proof of the optimality of the shortest remaining processing time discipline. In Operations Research, vol. 16, no. 3, pp. 678-690.
  12. Smith, D., 1976. A new proof of the optimality of the shortest remaining processing time discipline. In Operations Research, vol. 26, no. 1, pp. 197-199.
  13. Goerg, C., 1986. Evaluation of the optimal SRPT strategy with overhead. In IEEE Transactions on Communications, vol. 34, pp. 338-344.
  14. Harchol-Balter, M., Schroeder, B., Agrawal, M., Bansal, N., 2003. Size-based scheduling to improve web performance. In ACM Transactions on Computer Systems (TOCS), vol. 21, no. 2, pp. 207-233.
  15. Schroeder, B., Harchol-Balter, M., 2006. Web servers under overload: How schedule can help. In ACM TOIT, vol. 6, no. 1, pp. 20-52.
  16. Schroeder, B., Harchol-Balter, M., 2006. Web servers under overload: How schedule can help. In ACM TOIT, vol. 6, no. 1, pp. 20-52.
  17. Gong, M., Williamson, C., 2003. Quantifying the properties of SRPT scheduling. In MASCOTS, pp. 126-135.
  18. Friedman, E., Henderson, S., 2003. Fairness and efficiency in web server protocols. In ACM SIGMETRICS, pp. 229-237.
  19. Lu, D., Sheng, H., 2005. Effects and implications of file size/service time correlation on web server scheduling policies. In MASCOTS, pp. 258-267.
  20. Bansal, N., 2003. On the average sojourn time under M/M/1 SRPT. In ACM SIGMETRICS Performance Evaluation Review, vol. 31, no. 2, pp. 34-35.
  21. Bansal, N., Gamarnik, D., 2006. Handling load with less stress. In Queueing Systems, vol. 54 , no. 1, pp. 45-54.
  22. Barford, P., Crovella, M., 1998. Generating representative web workloads for network and server performance evaluation. In ACM joint international conference on Measurement and modeling of computer systems, pp. 151-160.
  23. Linux Foundation, 2007. Network Emulation (Netem). <http://www.linux-foundation.org/en/Net:Netem>
  24. Karn, P., Partridge, C., 1995. Improving round-trip time estimates in reliable transport protocols. In ACM SIGCOMM Computer Communication Review, vol. 25, no. 1, pp. 66-74.
  25. Jacobson, V., 1995. Congestion avoidance and Control. In ACM SIGCOMM Computer Communication Review, vol. 25, no.1, pp. 157-187.
  26. Network Services & Consulting Corporation, 2007. Internet Traffic Report. <http://www.internettraffic creport.com>
  27. Padhye, J., Firoiu, V., Towsley, D., Kurose, J., 2000. Modeling tcp reno performance: A simple model and its empirical validation. In IEEE/ACM Transactions on Networking (TON), vol. 8, no. 2, pp133-145.
  28. Crovella, M., Taqqu, M., Bestavros, A., 1998. Heavytailed probability distributions in the World Wide Web. In A Practical Guide To Heavy Tails, pages 3- 26. Chapman & Hall, New York.
  29. IRCache Home, 2004. The trace files. <ttp://www.ircache.net/Traces>
  30. Netcraft, 2007. Internet monitoring company. <http://news.netcraft.com/>
Download


Paper Citation


in Harvard Style

S. Al Sa'deh A. and H. Yahya A. (2008). IMPLEMENTATION OF A NEW SCHEDULING POLICY IN WEB SERVERS . In Proceedings of the Fourth International Conference on Web Information Systems and Technologies - Volume 1: WEBIST, ISBN 978-989-8111-26-5, pages 22-29. DOI: 10.5220/0001524200220029


in Bibtex Style

@conference{webist08,
author={Ahmad S. Al Sa'deh and Adnan H. Yahya},
title={IMPLEMENTATION OF A NEW SCHEDULING POLICY IN WEB SERVERS},
booktitle={Proceedings of the Fourth International Conference on Web Information Systems and Technologies - Volume 1: WEBIST,},
year={2008},
pages={22-29},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0001524200220029},
isbn={978-989-8111-26-5},
}


in EndNote Style

TY - CONF
JO - Proceedings of the Fourth International Conference on Web Information Systems and Technologies - Volume 1: WEBIST,
TI - IMPLEMENTATION OF A NEW SCHEDULING POLICY IN WEB SERVERS
SN - 978-989-8111-26-5
AU - S. Al Sa'deh A.
AU - H. Yahya A.
PY - 2008
SP - 22
EP - 29
DO - 10.5220/0001524200220029