Presentation 2014-11-13
Analyzing Log Data for Diagnosing Large Network Problems
Tatsuaki KIMURA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Network logs, such as router syslog, contain much useful information for daily health check and trouble shooting, however, they are not fully utilised due to its large volume and high diversity. In this work, we present methods of analyzing complicated log data in large scale network for diagnosing the network problems. First, we explain template extraction method to automatically extract primary templates from massive amount of unstructured log messages. Templates enable us to use complicated log messages in a uniform manner. In addition, by using templates, we present a log factorization technique, which models the generation of log data and factorizes them into meaningful network events. Our main idea is to consider the generation of log data as a rank-3 tensor and to apply non-negative tensor factorization approach. We then validate our techniques using massive amount of network log messages collected from a large operating network. We also demonstrate several case studies that validate the usefulness of our technique.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Network Management / Syslog / Nonnegative Tensor Factorization (NTF)
Paper # NS2014-131,CQ2014-74,ICM2014-22
Date of Issue

Conference Information
Committee ICM
Conference Date 2014/11/6(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Information and Communication Management(ICM)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Analyzing Log Data for Diagnosing Large Network Problems
Sub Title (in English)
Keyword(1) Network Management
Keyword(2) Syslog
Keyword(3) Nonnegative Tensor Factorization (NTF)
1st Author's Name Tatsuaki KIMURA
1st Author's Affiliation NTT Network Technology Laboratories, NTT Corporation()
Date 2014-11-13
Paper # NS2014-131,CQ2014-74,ICM2014-22
Volume (vol) vol.114
Number (no) 299
Page pp.pp.-
#Pages 6
Date of Issue