sas 15(1): e3

Research Article

Evaluating Unsupervised Fault Detection in Self-Healing Systems Using Stochastic Primitives

Download529 downloads
  • @ARTICLE{10.4108/sas.1.1.e3,
        author={Chris Schneider and Adam Barker and Simon Dobson},
        title={Evaluating Unsupervised Fault Detection in Self-Healing Systems Using Stochastic Primitives},
        journal={EAI Endorsed Transactions on Self-Adaptive Systems},
        volume={1},
        number={1},
        publisher={ICST},
        journal_a={SAS},
        year={2015},
        month={1},
        keywords={Self-healing; Systems; Fault; Anomaly; Detection; Machine Learning; Computational Intelligence; Autonomic Computing; Artificial Neural Networks; Self-Organising Maps; Hidden Markov Models; Restricted Boltzmann Machines; Recurrent Neural Networks.},
        doi={10.4108/sas.1.1.e3}
    }
    
  • Chris Schneider
    Adam Barker
    Simon Dobson
    Year: 2015
    Evaluating Unsupervised Fault Detection in Self-Healing Systems Using Stochastic Primitives
    SAS
    ICST
    DOI: 10.4108/sas.1.1.e3
Chris Schneider1,*, Adam Barker1, Simon Dobson1
  • 1: School of Computer Science, University of St Andrews, Fife, Scotland, KY16 9SX, UK
*Contact email: chris.schneider@st-andrews.ac.uk

Abstract

Autonomous fault detection represents one approach for reducing operational costs in large-scale computing environments. However, little empirical evidence exists regarding the implementation or comparison of such methodologies, or offers proof that such approaches reduce costs. This paper compares the effectiveness of several types of stochastic primitives using unsupervised learning to heuristically determine the root causes of faults. The results suggest that self-healing systems frameworks leveraging these techniques can reliably and autonomously determine the source of an anomaly within as little as five minutes. This finding lays the foundation for determining the potential these approaches have for reducing operational costs and ultimately concludes with new avenues for exploring anomaly prediction.