Proactive Fault Detection Schema for Enterprise Information System Using Statistical Process Control

Authors:
Chihoon Lee;Doohyung Lee;Jahwan Koo;Jinwook Chung
Affiliations:
School of Information and Communication Engineering, Sungkyunkwan University, Suwon, South Korea 440-746;School of Information and Communication Engineering, Sungkyunkwan University, Suwon, South Korea 440-746;Computer Sciences Department, University of Wisconsin-Madison, Madison, USA WI 53706;School of Information and Communication Engineering, Sungkyunkwan University, Suwon, South Korea 440-746
Venue:
Proceedings of the Symposium on Human Interface 2009 on ConferenceUniversal Access in Human-Computer Interaction. Part I: Held as Part of HCI International 2009
Year:
2009

Citing 1
Cited 1

Graph identification techniques applied to network management problems

Graph identification techniques applied to network management problems

Compiler-support for robust multi-core computing

ISoLA'10 Proceedings of the 4th international conference on Leveraging applications of formal methods, verification, and validation - Volume Part I

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper proposes a proactive fault detection schema using adaptive statistical approaches in order to enhance system availability and reliability in the heterogeneous & complicated information system environment. The proposed system applies Six Sigma SPC (Statistical Process Control) techniques already validated in industries in order to monitor the application system in the information system. This makes it possible to reduce false alarm rates for system faults and accurately detect faults by creating a control chart based on past performance data and controlling the distribution of performance based on the chart. The early detection of faults is also enabled through a fault prediction model. Therefore, the aforementioned system not only detect unknown or unseen faults but also resolve potential problems for system administrator by detecting abnormal behaviors before faults occur. In the experiment we show the superiority of our proposed model and the possibility to early detect system faults.