网络故障检测中的探测路径选择方法研究

doi:10.13232/j.cnki.jnju.2022.02.015

南京大学学报(自然科学版) ›› 2022, Vol. 58 ›› Issue (2): 320–327.doi: 10.13232/j.cnki.jnju.2022.02.015

• • 上一篇

网络故障检测中的探测路径选择方法研究

齐小刚, 马文超(), 李家慧

西安电子科技大学数学与统计学院，西安，710126

收稿日期:2021-11-22 出版日期:2022-03-30 发布日期:2022-04-02
通讯作者: 马文超 E-mail:15054438025@163.com
作者简介:E⁃mail：15054438025@163.com
基金资助:
国家自然科学基金(61877067);数据链技术重点实验室基金(CLDL?20182115);近地面探测与感知技术重点实验室基金(TCGZ2019A002);基础研究项目(61424140502)

The study of detection path selection methods in network fault detection

Xiaogang Qi, Wenchao Ma(), Jiahui Li

School of Mathematics and Statistics，Xidian University，Xi'an，710126，China

Received:2021-11-22 Online:2022-03-30 Published:2022-04-02
Contact: Wenchao Ma E-mail:15054438025@163.com

摘要/Abstract

摘要：

通信网络中数据传输能力强的节点实时负载高、传输价值高，在进行故障探测时会产生较高的探测成本.为了减少探测成本，提出一种基于主动探测的探测路径选择算法，该算法定义节点权值以衡量节点的数据传输能力.在探测站选择阶段，算法迭代地选择权值最小的节点作为探测站；在选取探针时，通过合适的 $K$ 值来限制探针长度，减少探针往返时间.算法在确保网络中所有节点都被探测到的情况下，选择满足条件的探针，扩大节点覆盖范围，以减少探针数量，降低探测成本.随机网络拓扑和真实网络拓扑的仿真结果表明，提出的故障检测算法和其他算法相比，能有效地减少探针数量和降低探测成本.

关键词: 故障检测, 主动探测, 探测成本, 传输价值, 通信网络

Abstract:

An important objective in existing detection?based network monitoring methods is to reduce the probing costs. Low probing costs mean low resource consumption and less negative impact on the transmission of data. Nodes with high data transmission capacity in the communication network have high real?time loads and high transmission value，therefore incur higher probing costs when performing fault detection. In this paper，a detection path selection algorithm based on active detection is proposed on the issue of reducing probing costs. The node weights are used to measure the data transmission capability of the nodes. In each iteration，the node with the lowest weight is selected as the probing station，which avoids the situation where detection packets are lost at nodes with large weights to lead to inaccurate detection results. When selecting probes，the probe length is limited by a suitable K value to reduce the probe round trip time. The algorithm expands node coverage by selecting suitable probes to ensure that all nodes in the network are detected. It reduces the number of probes and probing costs. Simulation results of random and real network topologies show that the fault detection algorithm proposed in this paper is effective in reducing the number of probes and decreasing probing costs compared to other algorithms.

Key words: fault detection, active detection, detection costs, transmission values, communication network

中图分类号:

TP311

齐小刚, 马文超, 李家慧. 网络故障检测中的探测路径选择方法研究[J]. 南京大学学报(自然科学版), 2022, 58(2): 320–327.

Xiaogang Qi, Wenchao Ma, Jiahui Li. The study of detection path selection methods in network fault detection[J]. Journal of Nanjing University(Natural Sciences), 2022, 58(2): 320–327.

图/表 12

图1

图2

图3

表1

候选探针集"

$c p 61$	$6,5, 4,3$
$c p 62$	$6,7, 2,1$
$c p 63$	$6,5, 4$
$c p 64$	$6,7, 2$
$c p 65$	$6,7, 8$
$c p 66$	$6,5$
$c p 67$	$6,7$

表1

图4

表2

表3

表4

表5

图5

图6

图7

参考文献 20

1	Tayal A， Sharma N， Hubballi N，et al. Traffic dynamics?aware probe selection for fault detection in networks. Journal of Network and Systems Management，2020，28(4)：1055-1084.
2	Wang S S， Xu H L， Huang L S，et al. Fast recovery for single link failure with segment routing in SDNs∥2019 IEEE 21st International Conference on High Performance Computing and Communications，IEEE 17th International Conference on Smart City，IEEE 5th International Conference on Data Science and Systems. Zhangjiajie，China：IEEE，2019：2013-2018.
3	Zheng J Q， Xu H， Zhu X J，et al. Sentinel：Failure recovery in centralized traffic engineering. IEEE/ACM Transactions on Networking，2019，27(5)：1859-1872.
4	Bogle J， Bhatia N， Ghobadi M，et al. TEAVAR：Striking the right utilization?availability balance in wan traffic engineering∥Proceedings of the ACM Special Interest Group on Data Communication. Beijing，China：ACM，2019：29-43.
5	Cheng L， Qiu X S， Meng L M，et al. Efficient active probing for fault diagnosis in large scale and noisy networks∥2010 Proceedings IEEE INFOCOM. San Diego，CA，USA：IEEE，2010：1-9.
6	Natu M， Sethi A S， Lloyd E L. Efficient probe selection algorithms for fault diagnosis. Telecommu?nication Systems，2008，37(1-3)：109-125.
7	Brodie M， Rish I， Ma S，et al. Active probing strategies for problem diagnosis in distributed systems∥Proceedings of the 18th International Joint Conference on Artificial Intelligence. Acapulco，Mexico：Morgan Kaufmann，2003：1337-1338.
8	Xuan Y， Shen Y L， Nguyen N P，et al. Efficient multi?link failure localization schemes in all?optical networks. IEEE Transactions on Communications，2013，61(3)：1144-1151.
9	Jeswani D， Natu M， Ghosh R K. Adaptive monitoring：Application of probing to adapt passive monitoring. Journal of Network and Systems Management，2015，23(4)：950-977.
10	Zhou H， Yang Y， Qiu X S，et al. The strategy of probe station selection of active probing in WSNs∥The 16^th Asia?Pacific Network Operations and Management Symposium. Hsinchu，Taiwan，China：IEEE，2014：1-4.
11	Patil B， Kinger S， Pathak V K. Probe station placement algorithm for probe set reduction in network fault localization∥2013 International Conference on Information Systems and Computer Networks. Mathura，India：IEEE，2013：164-169.
12	Jeswani D， Korde N， Patil D，et al. Probe station selection algorithms for fault management in computer networks∥2010 2nd International Conference on Communication Systems and NETworks. Bangalore，India：IEEE，2010：1-9.
13	Li H K， Gao Y， Dong W，et al. Taming both predictable and unpredictable link failures for network tomography. IEEE/ACM Transactions on Networking，2018，26(3)：1460-1473.
14	He T， Gkelias A， Ma L，et al. Robust and efficient monitor placement for network tomography in dynamic networks. IEEE/ACM Transactions on Networking，2017，25(3)：1732-1745.
15	Mardani M， Giannakis G B. Estimating traffic and anomaly maps via network tomography. IEEE/ACM Transactions on Networking，2016，24(3)：1533-1547.
16	Li H K， Gao Y， Dong W，et al. Preferential link tomography in dynamic networks. IEEE/ACM Transactions on Networking，2019，27(5)：1801-1814.
17	Ma L， He T， Leung K K，et al. Inferring link metrics from end?to?end path measurements：Identifiability and monitor placement. IEEE/ACM Transactions on Networking，2014，22(4)：1351-1368.
18	Feng C Y， Wang L N， Wu K，et al. Bound inference in network performance tomography with additive metrics. IEEE/ACM Transactions on Networking，2020，28(4)：1859-1871.
19	Geng H J， Zhang H， Shi X G，et al. A hybrid link protection scheme for ensuring network service availability in link?state routing networks. Journal of Communications and Networks，2020，22(1)：46-60.
20	Department of Computer Science & Engineering，University of Washington.Rocketfuel：An ISP topology mapping engine. .

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

Network	DPSA1	PDSA2	MaxFD	MinFD	RandomFD
(a)	3	4	4	4	4
(b)	4	4	5	3	4
(c)	4	5	4	4	4
(d)	4	5	4	5	5

AS	Nodes	Edges
3967	79	147
6461	141	374
3257	161	328
1239	315	972

AS	DPSA1	DPSA2	MaxFD	MinFD	RandomFD
3967	20	19	51	49	53
6461	31	34	57	63	68
3257	45	49	64	71	78
1239	53	57	161	153	179

[1]	伊小素,冀羽,曾华菘,熊瑞. 考虑邻接点贡献的通信网关键节点评估方法[J]. 南京大学学报(自然科学版), 2019, 55(5): 725-732.
[2]	李霞，杨文，苏阳 . 分布式多跳水声通信网络的媒质控制协议研究 [J]. 南京大学学报(自然科学版), 2013, 49(1): 72-77.

网络故障检测中的探测路径选择方法研究

The study of detection path selection methods in network fault detection

RichHTML

PDF (PC)

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献 20

相关文章 2

Metrics

本文评价

推荐阅读 0