Fault detection and localization in distributed systems using invariant relationships

@article{Sharma2013FaultDA,
  title={Fault detection and localization in distributed systems using invariant relationships},
  author={Abhishek B. Sharma and Haifeng Chen and Min Ding and Kenji Yoshihira and Guofei Jiang},
  journal={2013 43rd Annual IEEE/IFIP International Conference on Dependable Systems and Networks (DSN)},
  year={2013},
  pages={1-8}
}
Recent advances in sensing and communication technologies enable us to collect round-the-clock monitoring data from a wide-array of distributed systems including data centers, manufacturing plants, transportation networks, automobiles, etc. Often this data is in the form of time series collected from multiple sensors (hardware as well as software based). Previously, we developed a time-invariant relationships based approach that uses Auto-Regressive models with eXogenous input (ARX) to model… CONTINUE READING
Highly Cited
This paper has 24 citations. REVIEW CITATIONS

6 Figures & Tables

Extracted Numerical Results

  • This situation worsens if noise has high impact – the precision for nodeScore drops to 0.46 and 0.4 for 10% and 50% noisy broken invariants.
  • and spatial rank have reasonable precision even at 50% noisy broken invariants with spatial avg.
  • E.g. for a spike anomaly at t = 272, with noisy broken invariants, t = 275 has the highest anomalyScore (see Algorithm 2), and the precision of the temporal algorithm is 0.1.

Topics