Web servers under overload: How scheduling can help

被引:51
作者
Schroeder, Bianca [1 ,2 ]
Harchol-Balter, Mor [1 ]
机构
[1] Computer Science Dept., Carnegie Mellon University, Pittsburgh, PA 15213-3891
基金
美国国家科学基金会;
关键词
Overload; Scheduling; SRPT; Starvation; Unfairness; Web server;
D O I
10.1145/1125274.1125276
中图分类号
学科分类号
摘要
This article provides a detailed implementation study on the behavior of web serves that serve static requests where the load fluctuates over time (transient overload). Various external factors are considered, including WAN delays and losses and different client behavior models. We find that performance can be dramatically improved via a kernel-level modification to the web server to change the scheduling policy at the server from the standard FAIR (processor-sharing) scheduling to SRPT (shortest-remaining-processing-time) scheduling. We find that SRPT scheduling induces no penalties. In particular, throughput is not sacrificed and requests for long files experience only negligibly higher response times under SRPT than they did under the original FAIR scheduling. © 2006 ACM.
引用
收藏
页码:20 / 52
页数:32
相关论文
共 64 条
[21]  
Crovella M., Frangioso R., Harchol-Balter M., Connection scheduling in web servers, USENIX Symposium on Internet Technologies and Systems, OCT, (1999)
[22]  
Day M., Cain B., Tomlinson G., Rzewski P., A model for content internetworking (cdi), Internet Draft, MAY, (2002)
[23]  
Dias D.M., Kish W., Mukherjee R., Tewari R., A scalable and highly available web server, COMPCON, pp. 85-92, (1996)
[24]  
Druschel P., Banga G., Lazy receiver processing (LRP): A network subsystem architecture for server systems, Proceedings of OSDI '96, OCT, pp. 261-275, (1996)
[25]  
Elnikety S., Nahum E.M., Tracey J., Zwaenepoel W., A method for transparent admission control and request scheduling in dynamic e-commerce Web sites, International World-wide Web Conference (WWW'04), MAY, (2004)
[26]  
Feldmann A., Web performance characteristics, IETF, NOV
[27]  
Friedman E.J., Henderson S.G., Fairness and efficiency in web server protocols, Proceedings of the ACM SIGMETRICS Conference on Measurement and Modeling of Computer Systems, MAY, (2003)
[28]  
Gong M., Williamson C., Quantifying the properties of SRPT scheduling, Proceedings of IEEE /ACM International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS)
[29]  
Gwertzman J., Seltzer M., The case for geographical push-caching, Proceedings of HotOS '94, MAY, (1994)
[30]  
Harchol-Balter M., Schroeder B., Agrawal M., Bansal N., Size-based scheduling to improve web performance, ACM Trans. Comput. Syst., 21, 2 MAY, (2003)