DRACA: Decision-support for Root Cause Analysis and Change Impact Analysis

IT Management;Decision Support;Mining Software Repositories;Computer Science

Nadi, Sarah

University of Waterloo

关键词: IT Management;    Decision Support;    Mining Software Repositories;    Computer Science;   

Others  :  https://uwspace.uwaterloo.ca/bitstream/10012/4889/1/NADI_SARAH.pdf

瑞士|英语

来源: UWSPACE Waterloo Institutional Repository

【 摘 要 】

Most companies relying on an Information Technology (IT) system for theirdaily operations heavily invest in its maintenance. Tools that monitor networktraffic, record anomalies and keep track of the changes that occur in the systemare usually used. Root cause analysis and change impact analysis are two mainactivities involved in the management of IT systems. Currently, there exists nouniversal model to guide analysts while performing these activities. Although theInformation Technology Infrastructure Library (ITIL) provides a guide to the or-ganization and structure of the tools and processes used to manage IT systems, itdoes not provide any models that can be used to implement the required features.This thesis focuses on providing simple and effective models and processes forroot cause analysis and change impact analysis through mining useful artifactsstored in a Confguration Management Database (CMDB). The CMDB containsinformation about the different components in a system, called Confguration Items(CIs), as well as the relationships between them. Change reports and incidentreports are also stored in a CMDB. The result of our work is the Decision supportfor Root cause Analysis and Change impact Analysis (DRACA) framework whichsuggests possible root cause(s) of a problem, as well as possible CIs involved in a change set based on di erent proposed models. The contributions of this thesis areas follows:- An exploration of data repositories (CMDBs) that have not been previouslyattempted in the mining software repositories research community.- A causality model providing decision support for root cause analysis basedon this mined data.- A process for mining historical change information to suggest CIs for futurechange sets based on a ranking model. Support and con dence measures areused to make the suggestions.- Empirical results from applying the proposed change impact analysis processto industrial data. Our results show that the change sets in the CMDB werehighly predictive, and that with a confidence threshold of 80% and a halflife of 12 months, an overall recall of 69.8% and a precision of 88.5% wereachieved.- An overview of lessons learned from using a CMDB, and the observations wemade while working with the CMDB.