Presentation | 2022-10-05 A Feature Extraction Method From Command Outputs of Network Devices Keitaro Kaida, Tatsuaki Kimura, Hiroshi Yamauchi, Tetsuya Takine, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | With the diversification and rapid growth of communication services, monitoring and management of networks have become increasingly complex. Therefore, the conventional failure diagnosis procedure, in which network operators execute network commands to check the device status and to determine the failure locations and causes, does not scale, and its automation is in high demand. However, the output logs of these commands for troubleshooting, called device command outputs, are unstructured text messages consisting of tens to hundreds of lines, and therefore automatic analysis and information extraction have been challenging tasks. In this study, we propose a feature extraction method from device command outputs, which automatically identifies templates, i.e., primary parts of device command outputs, and extracts parameter variables that vary with command outputs as features. The proposed method first divides each command output into blocks, each of which constitutes a template. The blocks are then merged into clusters and a sequence of words that commonly appear within the blocks in each cluster is extracted as a template. Finally, parameter variables are identified by using the templates, and feature vectors are constructed by considering the types of words, such as numbers and symbols. Furthermore, we apply the proposed method to data obtained from a real network to evaluate the accuracy of template extraction. Moreover, as an example of the application of the extracted feature vectors, we perform anomaly detection using device command outputs and demonstrate the usefulness of the proposed method. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Network operations / failure diagnosis / device command output / feature extraction |
Paper # | NS2022-76 |
Date of Issue | 2022-09-28 (NS) |
Conference Information | |
Committee | NS |
---|---|
Conference Date | 2022/10/5(3days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Hokkaidou University + Online |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | Network architecture (5G, Local 5G, Beyond5G, Mobile networks, Ad-hoc and sensor networks, Overlay and P2P networks, Programmable networks, SDN/NFV, IoT, Network slicing), Next generation packet transport (High speed Ethernet, IP over WDM, Multi-service package technology, MPLS), Grid, etc. |
Chair | Tetsuya Oishi(NTT) |
Vice Chair | Takumi Miyoshi(Shibaura Insti of Tech.) |
Secretary | Takumi Miyoshi(NTT) |
Assistant | Kotaro Mihara(NTT) |
Paper Information | |
Registration To | Technical Committee on Network Systems |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | A Feature Extraction Method From Command Outputs of Network Devices |
Sub Title (in English) | |
Keyword(1) | Network operations |
Keyword(2) | failure diagnosis |
Keyword(3) | device command output |
Keyword(4) | feature extraction |
1st Author's Name | Keitaro Kaida |
1st Author's Affiliation | Osaka University(Osaka Univ.) |
2nd Author's Name | Tatsuaki Kimura |
2nd Author's Affiliation | Osaka University(Osaka Univ.) |
3rd Author's Name | Hiroshi Yamauchi |
3rd Author's Affiliation | Softbank Corporation(SBKK) |
4th Author's Name | Tetsuya Takine |
4th Author's Affiliation | Osaka University(Osaka Univ.) |
Date | 2022-10-05 |
Paper # | NS2022-76 |
Volume (vol) | vol.122 |
Number (no) | NS-198 |
Page | pp.pp.1-1(NS), |
#Pages | 1 |
Date of Issue | 2022-09-28 (NS) |