期刊文献+

基于响应时间的Web服务器性能评测方法 被引量:1

Towards to Evaluating Performance of Web Server Based on Response Time
下载PDF
导出
摘要 Web服务器的性能评测是Web服务容量规划和服务部署中的重要问题.针对目前方法在评测实际应用场景中的不足,根据峰值负载之前响应时间与用户请求量存在的指数关系,提出Web服务器性能评测模型MBRT(Web Server performance evaluation Model Based on Response Time),根据响应时间建模可以避免分析的不全面,有效的避免传统模型的局限性,然后基于MBRT提出具体评测方法,同时使用一元线性回归模型的显著性特征验证MBRT的准确性.最后为了显示MBRT的有效性和高效性,在Web服务器实际评测环境中对MBRT进行验证和评估,并与已有模型进行比较. Performance evaluation of Web Server plays an important role in in performance analysis,evaluation and capacity planning.It also can be used to control the overload of a Web server and guide the allocation of resources.As a result,there have been many works focused on evaluating the performance of Web servers by a special evaluation model effectively and accurately.These works can be mainly divided into two types: 1) Analysis/simulation based model,2) Measurement based model.Because many factors are not found in the simulated environment,and we cannot predict that which factors may been discovered in some real environment using the former.The latter often evaluates the performance of a Web server according to the utilization of resource,like CPU,memory and disk.However,resource is not the only one that affected the performance of a Web server.In order to solve the problems,we propose a Web Server performance evaluation Model Based on the Response Time-MBRT.The idea of model is the exponential relationship between Response Time and users′ requests before Web servers attach the peak load.This can ensure us analyze the real application in the round and avoid the limitations of traditional model.We also give the evaluation process using MBRT.Furthermore,the correctness of MBRT is verified with unary regression.At last,the tests are conducted in real applications.By comparing to some existing model,the test results show the effectiveness and efficiency of MBRT.
出处 《小型微型计算机系统》 CSCD 北大核心 2013年第1期90-95,共6页 Journal of Chinese Computer Systems
基金 国家自然科学基金项目(60973029)资助
关键词 WEB服务器性能 评测模型 峰值负载 performance of Web server evaluation model peak load
  • 相关文献

参考文献20

  • 1Guangzhu J, Shujuan J. A quick testing model of Web performance based on testing flow and its application [ C ]. Web Information Systems and Applications Conference,2009 : 57 -61.
  • 2Rak M, Aversa R, Mrtino B D, et al. Web services resilience e- valuation using LDS load dependent server models [ J_]. Journal of Communications,2010,5 ( 1 ) :39-49.
  • 3Fujii T, Dohi T. Statistical failure analysis of a Web server system [ C ]. International Conference on Availability, Reliability and Se- curity ,2009:554-559.
  • 4Ziya S. On the relationships among traffic load, capacity, and throughput for the M/M/l/m, queues [ J ]. IEEE Transactions ( 11 ) :2696-2701. M/G/1/m-PS, and M/G/c/c on Automatic Control, 2008,53.
  • 5lothouber L P. A model of web server performance [ M ]. Proceed- ings of the Fifth International World Wide Web, 1996.
  • 6Shivam P, Marupadi V, Chase J, et al. Cutting comers : work- bench automation for server benchmarking [ C ]. USENIX Annual Technical Conference,2008.
  • 7Yin J, Ming Z, Xiao Z, et al. A Web performance modeling process based on the methodology of learning form date [ C]. 9th International Conference for Young Computer Scientists, 2008: 1285-1291.
  • 8Welsh M, Culler D. Adaptive overload control for busy internet servers [ C]. 4th Conference on USENIX Symposium on lntemetTechnologies and Systems, 2003.
  • 9Doyle R P, Chase J S, Asad O M, et al. Model-based resource provisioning in a web service utility[ C]. 4th Conference on USE- NIX Symposium on Intemet Technologies and Systems ,2003.
  • 10Wei Zong-shu. Probability theory and mathematical statistics [ M]. Beijing : High Education Press ,2005.

同被引文献8

  • 1Liu B. Characterizing Web response time[D]. Blacksburg: Verginia Polytechnic Institute and State University, 1998.
  • 2On the relationships among traffic load, capacity, and throughput for the mmlm, mglm- ps, and mgcc queues[J]. IEEE transactions on automatic control, 2008,53 ( 11 ).
  • 3Chen Liang. Predicting Network Response Times Using Social Information [C]. 2011 International Conference on Advances in Social Networks Analysis and Mining, 2011:527-531.
  • 4Yuliy Baryshnikov. Predictability of Web-Server Traffic Congestion[C]. Proceedings of the 10th International Workshop on Web Content Caching and Distribution(WCW'05), 2005.
  • 5Xiaoying Zou. Analysis of user perceived web quality of service delay[C]. Procedings of IC- NIDC2012: 319-324.
  • 6浦云明,王宝玉.基于负载性能指标的Web测试[J].计算机系统应用,2010,19(5):220-223. 被引量:16
  • 7郑晓霞,赵俊峰,程志文,谢冰.一种WebService响应时间的动态预测方法[J].小型微型计算机系统,2011,32(8):1570-1574. 被引量:4
  • 8梁晟,李明树,梁金能,陈振冲.Web应用程序运行响应时间的实验研究[J].计算机研究与发展,2003,40(7):1076-1080. 被引量:12

引证文献1

二级引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部