Research on Network Performance Measurement and Optimization Technology in High Energy Physical Grid Environment

doi:10.11871/jfdc.issn.2096-742X.2021.05.007

Abstract

Abstract:

[Objective] We use the SDN architecture network to optimize the selection of high-energy physical data transmission paths, and improve the data transmission efficiency between high-energy physics cooperation units. [Methods] In this paper, we use the service-oriented network performance measurement framework perfSONAR, GRE-based virtual network technology, and SDN-based control technology to achieve our goals. [Results] We propose a method for network performance measurement and performance optimization in a high-energy physical grid environment. First, SDN technology and virtual network technology are used to construct a high-energy physical data sharing virtual private network based on the SDN architecture in the high-energy physical grid environment. Then, perfSONAR is used to design and implement a network performance measurement and visualization platform. Given the results of network performance measurement in the high-energy physics grid environment, the optimal data sharing path between two nodes can be selected based on the optimal path calculation method proposed in this paper. [Conclusions] Test results show that this method can improve the efficiency of data sharing, and provide a reliable guarantee for high-speed and stable data transmission between nodes in the high-energy physics grid environment.

Key words: network measurement, network performance, grid technology, SDN, virtual private network

SUN Zhihui,QI Fazhi. Research on Network Performance Measurement and Optimization Technology in High Energy Physical Grid Environment[J]. Frontiers of Data and Computing, 2021, 3(5): 98-108.

Figures/Tables 8

Fig.1

Fig.2

Fig.3

Fig.4

Table 1

Table 2

Table 3

Table 4

References 21

[1]	齐法制, 孙智慧, 姚辉, 等. 基于SDN的高能物理数据交换虚拟专用网研究[J]. 计算机工程与应用, 2016, 52(8),106-110.
[2]	程耀东, 石京燕, 陈刚. 高能物理计算环境概述[J]. 科研信息化技术与应用, 2014, 5(3):3-10.
[3]	OL: http://glite.web.cern.ch/glite .
[4]	胡治国, 田春岐, 杜亮, 等. IP网络性能测量研究现状和进展[J]. 软件学报, 2017, 28(1):105134.
[5]	Mohan V, Reddy YRJ, Kalpana K. Active and passive network measurements: A survey[J]. International Journal of Computer Science and Information Technologies, 2011, 2(4):1372-1385.
[6]	尹浩, 李峰. 互联网性能测量技术发展研究[J]. 计算机研究与发展, 2016, 53(1):3-14.
[7]	刘敏, 李忠诚, 过晓冰, 等. 端到端的可用带宽测量方法[J]. 软件学报, 2006, 17(1):108116.
[8]	Adnan A, authorRicky M, Zubair S. FlowTrace: A Framework for Active Bandwidth Measurements Using In-band Packet Trains[C]. The Int’l Conf. on Passive and Active Network Measurement, 2020,37-51.
[9]	Yu C, Lumezanu C, Sharma A, et al. Software-defined latency monitoring in data center networks[C]. The Int’l Conf. on Passive and Active Network Measurement, 2015: 360-372.
[10]	Sinha D, Haribabu K, Balasubramaniam S. Real-time monitoring of network latency in software defined networks[C]. In:Proceedings of IEEE Inter-national Conference on Advanced Networks and Telecom-muncations Systems, Kolkata, 2015:1-3.
[11]	Adrichem N L M V, Doerr C, Kuipers F A. OpenNetMon: network monitoring in OpenFlow software-defined networks[C]. In: Proceedings of IEEE Network Opera-tions and Management Symposium, Krakow, 2014:1-8.
[12]	张广兴, 邱峰, 谢高岗, 等. 一种高效的网络流记录表示方法[J]. 计算机研究与发展, 2013, 50(4):722-730.
[13]	徐鹏, 刘琼, 林森. 基于支持向量机的Internet流量分类研究[J]. 计算机研究与发展, 2009, 46(3):407-414.
[14]	王风宇, 郭山清, 李亮雄, 等. 一种高效率的大流提取方法[J]. 计算机研究与发展, 2013, 50(4):731-740.
[15]	Sanjuàs-Cuxart J, Barlet-Ros P, Duffield N, et al. Cuckoo sampling: Robust collection of flow aggregates under a fixed memory budget[C]. In: Proc. of the 31st Annual IEEE Int’l Conf. on Computer Communications (Mini-Conf.). 2012: 2751-2755.
[16]	Ma XY, Hu CC, Jiang JC, et al. S3: Smart selection of sampling function for passive network measurement[C]. In: Proc. of the 36th Annual IEEE Conf. on Local Computer Networks. Bonn: IEEE, 2011: 416-423.
[17]	John W, Meirosu C. Low-overhead packet loss and one-way delay measurements in service provider SDN[C]. ONS, 2014:1-2.
[18]	Sezer S, Scott-Hayward S, Chouhan P K, et al. Are we ready for SDN? implementation challenges for software-defined networks[J]. IEEE Commun Mag, 2013, 51:36-43.
[19]	王阿领, 刘渊, 王晓锋. 主被动结合的网络测量技术研究[J]. 计算机工程与应用, 2020, 56(21):93-102.
[20]	Zangrilli M, Lowekamp B B. Applying principles of active available bandwidth algorithms to passive TCP traces[C]. Passive and Active Network Measurement. Berlin: Springer, 2005:333-336.
[21]	Vaibhav Bajpai; Jürgen Schönwälder. A Survey on Inter-net Performance Measurement Platforms and Related Standardization Efforts[J]. IEEE Communications Surveys & Tutorials( Volume: 17, Issue: 3, thirdquarter 2015: 1313-1341.

链路描述	IPv4	IPv6	Cost_^g(ab)	Cost_v6(ab)
IHEP->SJTU	487.2	799.6	1	1
SJTU->IHEP	373.7	783.3	3	2
IHEP->SDU	324.6	756.7	4	3
SDU->IHEP	381.3	708.3	2	4
SJTU->SDU	313.7	776	5	6
SDU->SJTU	266.8	573	6	5

链路描述	IPv4	IPv6	Cost_^g(ab)	Cost_v6(ab)
IHEP->SJTU	26.74	26.33	6	5
SJTU->IHEP	26.68	26.36	5	6
IHEP->SDU	16.39	15.96	1	1
SDU->IHEP	16.43	16.13	2	2
SJTU->SDU	26.16	25.79	4	4
SDU->SJTU	26.07	25.51	3	3

链路描述	IPv4	IPv6	Cost_^g(ab)	Cost_v6(ab)
IHEP->SJTU	13	19	1	1
SJTU->IHEP	27	39	3	4
IHEP->SDU	29	31	4	2
SDU->IHEP	46	49	5	5
SJTU->SDU	23	37	2	3
SDU->SJTU	89	63	6	5

链路描述	cost1	cost2	cost3
IHEP->SDU IPv4	3.7	3.7	3.7
IHEP->SDU IPv6	2.4	2.5	2.6
IHEPU->SJTU->SDU only IPv4	5.2	5.5	5.8
IHEP->SJTU->SDU only IPv6	5	4.7	5.2
IHEP->SJTU->SDU IPv4->IPv6	5.1	5.2	5.3
IHEP->SJTU->SDU IPv6->IPv4	5.1	5.4	5.7