Performance evaluation method based on objective weight determination for data center network

doi:10.11772/j.issn.1001-9081.2015.11.3055

Journal of Computer Applications ›› 2015, Vol. 35 ›› Issue (11): 3055-3058.DOI: 10.11772/j.issn.1001-9081.2015.11.3055

• DPCS 2015 Paper • Previous Articles Next Articles

Performance evaluation method based on objective weight determination for data center network

NAN Yang, CHEN Lin

School of Computer, National University of Defense Technology, Changsha Hunan 410073, China

Received:2015-06-17 Revised:2015-07-15 Online:2015-11-10 Published:2015-11-13

基于客观权重确定的数据中心网络性能评估方法

南洋, 陈琳

国防科学技术大学计算机学院, 长沙 410073

通讯作者: 南洋(1987-),男,河南新乡人,硕士研究生,主要研究方向:网络管理、网络故障预警.
作者简介:陈琳(1976-),女,福建陇海人,副教授,博士,主要研究方向:网络管理、故障诊断、网络测量.
基金资助:
国家863计划项目(2012AA01A50606).

Abstract

Abstract: For large-scale data center network, how to monitor the network effectively, discover the bottleneck of network performance and potential point of failure, provide support for optimization of network performance becomes the new research subject. However, there are many factors which affect the network performance, and there are differences in the influence of performance factors. How to give an accurate performance evaluation has been a difficult problem. To solve these problems, a network performance evaluation index system was proposed in this paper. On this basis, a method for evaluating the network performance of data centers based on objective weight (PE-OWD) was put forward. By using the method of objective weight determination, the dynamic calculation of the weights of the indexes was adopted. And using the data normalization method based on historical parameters, a perfect network performance evaluation model was established. For the Tianhe2 real network environment, the performance indexes of the network equipment were evaluated, and the validity of the method for evaluating the network performance was verified.

Key words: performance evaluation, objective weights determination, health value, performance index

摘要： 针对大规模数据中心网络如何有效监控网络、发现网络性能瓶颈和潜在故障点,为网络性能优化提供支持成为新的研究课题.然而影响网络性能的因素众多,性能因素的影响程度存在差异,如何给出一个准确的性能评估一直是比较困难的问题.针对上述问题,提出了网络性能评估指标体系,在此基础上进一步提出了一种基于客观权重确定的数据中心网络性能评估方法(PE-OWD).该方法通过采用基于客观权值确定方法动态计算性能权值,利用基于历史参数分布的数据归一化方法,建立了完善的网络性能健康度评估模型.针对天河2真实的网络环境,对网络设备的性能指标进行评估,验证了网络性能评估方法的有效性.

关键词: 性能评估, 客观权重确定, 健康值, 性能指标

CLC Number:

TP393.07

NAN Yang, CHEN Lin. Performance evaluation method based on objective weight determination for data center network[J]. Journal of Computer Applications, 2015, 35(11): 3055-3058.

南洋, 陈琳. 基于客观权重确定的数据中心网络性能评估方法[J]. 计算机应用, 2015, 35(11): 3055-3058.

References

[1] MYSORE R N, PAMBORIS A, FARRINGTON N, et al. PortLand: a scalable fault-tolerant layer 2 data center network fabric[J]. ACM SIGCOMM Computer Communication Review, 2009, 39(4):39-50.
[2] GREENBERG A, HAMILTON J R, JAIN N, et al. VL2: a scalable and flexible data center network[J]. ACM SIGCOMM Computer Communication Review, 2009, 39(4):95-104.
[3] GUO C, WU H, TAN K, et al. DCell: a scalable and fault-tolerant network structure for data centers[C]// Proceedings of the 2008 ACM SIGCOMM. New York: ACM, 2008, 38(4):75-86.
[4] TURNER D, LEVCHENKO K, SNOEREN A C, et al. California fault lines: understanding the causes and impact of network failures[J]. ACM SIGCOMM Computer Communication Review, 2010, 40(4):315-326.
[5] WU X, TURNER D, CHEN C C, et al. NetPilot: automating datacenter network failure mitigation[J]. ACM SIGCOMM Computer Communication Review, 2012, 42(4):419-430.
[6] TANG F, TANG H, QIAHAN H, et al. A network performance assessment method based on passive measurement[J]. Computer and Digital Engineering, 2013, 282(4): 602-604.(唐飞,唐海娜,恰汗·合孜尔,等.一种基于被动测量的网络性能评估方法[J].计算机与数字工程,2013, 282(4): 602-604.)
[7] MENG X, MENG X. Network health evaluation based on SVM and cloud model[J]. Journal of Beijing University of Posts and Telecommunications, 2012,35(1):10-14.(温祥西,孟相如.基于支持向量机和云模型的网络健康状态评估[J].北京邮电大学学报,2012,35(1):10-14.)
[8] TONG X. A network based estimation method of business system[J]. Journal of Chongqing University of Technology: Natural Science, 2012,26(8):101-105,126.(童晓薇.一种基于网络的业务系统健康度评估方法[J].重庆理工大学学报:自然科学版,2012,26(8):101-105,126.)
[9] MA L.Study and use of modern statistical analysis methods (the third): the standardization of statistical data-dimensionless method[J]. Beijing Statistics, 2000(3):34-35.(马立平. 现代统计分析方法的学与用(三)∶统计数据标准化——无量纲化方法[J]. 北京统计, 2000(3):34-35.)
[10] WANG L, ZHANG R, SHA C, et al. A product normalization method for E-commerce [J]. Chinese Journal of Computers, 2014,37(2):312-325.(王立,张蓉,沙朝锋,等.电子商务商品归一化方法研究[J].计算机学报,2014,37(2):312-325.)
[11] YUAN H. Intelligent network management system: research on network performance evaluation model based on BP algorithm [D].Guangzhou: Guangdong University of Technology, 2006: 33-50.(原慧琴.智能网络管理系统——基于BP算法的网络性能评估模型的研究[D].广州: 广东工业大学, 2006: 33-50.)
[12] CAO W, CHEN G, NIU G,et al. A summarization on technologies of network device test[J]. Electronics Optics and Control, 2014,21(6):13-18.(曹文斌,陈国顺,牛刚,等.网络设备测试技术综述[J].电光与控制,2014,21(6):13-18.)
[13] LUO Y, XIA J, CHEN T. Comparison of objective weight determination methods in network performance evaluation [J]. Journal of Computer Applications, 2009, 29(10):2624-2626.(罗赟骞, 夏靖波, 陈天平. 网络性能评估中客观权重确定方法比较[J].计算机应用, 2009, 29(10):2624-2626.)

[1]	ZHU Chunmei, MO Hongqiang. Encoding of genetic algorithm for a class of fitness functions [J]. Journal of Computer Applications, 2017, 37(7): 1972-1976.
[2]	LI Ani, ZHANG Xiao, ZHANG Boyang, LIU Chunyi, ZHAO Xiaonan. Research on performance evaluation method of public cloud storage system [J]. Journal of Computer Applications, 2017, 37(5): 1229-1235.
[3]	YUAN Peiyan, WANG Chenyang, LI Sijia. Data forwarding strategy based on gathering point in mobile opportunistic networks [J]. Journal of Computer Applications, 2015, 35(11): 3038-3042.
[4]	XU Chunlei LI Junfeng QIU Yuan XIA Risheng YAN Yonghong. Parameters design and optimization of crosstalk cancellation system for two loudspeaker configuration [J]. Journal of Computer Applications, 2014, 34(5): 1503-1506.
[5]	ZHAO Huaqiong TANG Xuewen. Evaluation model of network service performance based on fuzzy analytic hierarchy process [J]. Journal of Computer Applications, 2013, 33(11): 3035-3038.
[6]	CHEN Chao-fuCHEN WANG Lei. Network time protocol performance evaluation in LAN environment [J]. Journal of Computer Applications, 2012, 32(04): 943-945.
[7]	WANG Jian-jiang QIU Di-shan PENG Li. Performance evaluation of space information data processing system based on queuing network [J]. Journal of Computer Applications, 2012, 32(03): 870-873.
[8]	. Design and implementation of cost measurement for thread context switch [J]. Journal of Computer Applications, 2010, 30(8): 2052-2055.
[9]	. Nonlinear dimensionality reduction algorithm and application to hospital performance evaluation [J]. Journal of Computer Applications, 2010, 30(4): 1004-1007.
[10]	. Implementation of hardware-in-loop network simulation platform based on NS-2 [J]. Journal of Computer Applications, 2010, 30(12): 3284-3287.
[11]	Gang Jiang. Weighted multi-step chaotic method of deep pump impeller ablation index in sandiness water [J]. Journal of Computer Applications, 2010, 30(11): 3101-3104.
[12]	. Comparison of objective weight determination methods in network performance evaluation [J]. Journal of Computer Applications, 2009, 29(10): 2624-2626.
[13]	. RTTM: New interconnection network stucture with hierarchical twist Torus [J]. Journal of Computer Applications, 2009, 29(08): 2149-2152.
[14]	XU Xiang WU jin-zhao Lin lian-nan Chen Jian-feng. Model checking interactive Markov chains [J]. Journal of Computer Applications, 2008, 28(7): 1868-1871.
[15]	Jun Hu Xian-hai Tan Yu-qing Hu Xue Yang. Reconstruction model of different network traffic based on multi-fractal [J]. Journal of Computer Applications, 2008, 28(3): 586-587.

Performance evaluation method based on objective weight determination for data center network

基于客观权重确定的数据中心网络性能评估方法

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics