基于排队论的在线交易处理服务器配置决策方案
Online Transaction Processing Server Configuration Decision Scheme Based on Queuing Theory
DOI: 10.12677/CSA.2020.1010186, PDF,    国家自然科学基金支持
作者: 张 乐*, 曾国荪#：同济大学计算机科学与技术系，上海；连增申：北京捷软世纪信息技术有限公司，北京；丁春玲：同济大学化学科学与工程学院，上海
关键词: 在线交易；服务器扩容；排队论；硬件性能配置；Online Transaction； Server Expansion； Queuing Theory； Hardware Performance Configuration
摘要: 电子商务频繁，在线交易量指数式增长，要求网络处理平台不断扩容升级。针对在线交易中心，服务器配置不准确，适应性差的问题，本文研究一种基于排队论的服务器配置方法。首先，采用G/M/c的排队论模型对交易请求的处理过程进行仿真建模，给出交易请求响应时间的估算公式，由此设计交易服务速度的计算方法。然后，通过实验建立服务器服务速度与服务器物理配置之间的对应关系，给出服务器物理配置的决策方案。实际应用表明，本文方法相可满足各种在线交易的服务器集群配置需求。
Abstract: Frequent e-commerce and exponential growth in online transaction volume require continuous expansion and upgrade of network processing platforms. Aiming at the problem of inaccurate server configuration and poor adaptability in online trading centers, this paper studies a server configuration method based on queuing theory. First, the G/M/c queuing theory model is used to simulate the processing process of the transaction request, and the estimating formula of the transaction request response time is given, thereby designing the calculation method of the transaction service speed. Then, the corresponding relationship between server service speed and server physical configuration is established through experiments, and a decision-making plan for server physical configuration is given. Practical application shows that the method in this paper can meet the server cluster configuration requirements of online transactions in various Chinese companies.
文章引用：张乐, 连增申, 曾国荪, 丁春玲. 基于排队论的在线交易处理服务器配置决策方案[J]. 计算机科学与应用, 2020, 10(10): 1753-1764. https://doi.org/10.12677/CSA.2020.1010186

参考文献

[1] 刘文清. 应用系统服务器的配置估算与调整[C]//国网信息通信有限公司、中国电机工程学会电力信息化专委会. 2008电力行业信息化年会会议论文集. 2008: 4.
[2] 陈英达, 钟苏生, 周开东, 等. 信息系统资源估算方法研究[J]. 信息技术与信息化, 2017(11): 133-136.
[3] Doran, D., Lipsky, L. and Thompson, S. (2010) Cost-Based Opti-mization of Buffer Size in M/G/1/N Systems under Different Service-Time Distributions. Ninth IEEE International Symposium on Network Computing and Applications, Cambridge, 15-17 July 2010, 28-35. [Google Scholar] [CrossRef] 
[4] Lee, Y.C., Wang, C., Zomaya, A.Y., et al. (2012) Profit-Driven Sched-uling for Cloud Services with Data Access Awareness. Journal of Parallel and Distributed Computing, 72, 591-602. [Google Scholar] [CrossRef] 
[5] Khazaei, H., Misic, J. and Misic, V.B. (2011) Performance Analy-sis of Cloud Computing Centers Using m/g/m/m + r Queuing Systems. IEEE Transactions on Parallel and Distributed Systems, 23, 936-943. [Google Scholar] [CrossRef] 
[6] Cao, J., Hwang, K., Li, K., et al. (2012) Optimal Multiserver Config-uration for Profit Maximization in Cloud Computing. IEEE Transactions on Parallel and Distributed Systems, 24, 1087-1096. [Google Scholar] [CrossRef] 
[7] Chiang, Y.J. and Ouyang, Y.C. (2014) Profit Optimiza-tion in SLA-Aware Cloud Services with a Finite Capacity Queuing Model. Mathematical Problems in Engineering, 2014, Article ID: 534510. [Google Scholar] [CrossRef] 
[8] Mei, J., Li, K., Ouyang, A., et al. (2015) A Profit Maximization Scheme with Guaranteed Quality of Service in Cloud Computing. IEEE Transactions on Computers, 64, 3064-3078. [Google Scholar] [CrossRef] 
[9] 廖倩文, 潘久辉, 王开杰. 基于排队理论的云计算中心性能分析模型[J]. 计算机工程, 2015, 41(9): 51-55.
[10] 张江强, 赵宁, 刘文奇. 具有两类请求的云计算中心服务器数量的优化[J]. 智能系统学报, 2017, 12(5): 601-607.
[11] 孙荣恒. 排队论基础[M]. 北京: 科学出版社, 2002.
[12] Yeh, K.C. and Kwan, K.C. (1978) A Comparison of Numerical Integrating Algorithms by Trapezoidal, La-grange, and Spline Approximation. Journal of Pharmacokinetics & Biopharmaceutics, 6, 79-98. [Google Scholar] [CrossRef] 
[13] 从树. 服务器硬件选型测试平台的设计与实现[D]: [硕士学位论文]. 武汉: 华中师范大学, 2017.
[14] 孙大为, 张广艳, 郑纬民. 大数据流式计算: 关键技术及系统实例[J]. 软件学报, 2014, 25(4): 839-862.
[15] 刘慧婷, 张旻, 程家兴. 基于多项式拟合算法的EMD端点问题的处理[J]. 计算机工程与应用, 2004(16): 84-86 + 100.