Being Accurate Is Not Enough: New Metrics for Disk Failure Prediction

@article{Li2016BeingAI,
  title={Being Accurate Is Not Enough: New Metrics for Disk Failure Prediction},
  author={J. Li and R. Stones and G. Wang and Zhongwei Li and X. Liu and K. Xiao},
  journal={2016 IEEE 35th Symposium on Reliable Distributed Systems (SRDS)},
  year={2016},
  pages={71-80}
}
  • J. Li, R. Stones, +3 authors K. Xiao
  • Published 2016
  • Computer Science
  • 2016 IEEE 35th Symposium on Reliable Distributed Systems (SRDS)
  • Traditionally, disk failure prediction accuracy is used to evaluate disk failure prediction model. However, accuracy may not reflect their practical usage (protecting against failures, rather than only predicting failures) in cloud storage systems. In this paper, we propose two new metrics for disk failure prediction models: migration rate, which measures how much at-risk data is protected as a result of correct failure predictions, and mismigration rate, which measures how much data is… CONTINUE READING
    25 Citations
    New Metrics for Disk Failure Prediction That Go Beyond Prediction Accuracy
    • 3
    Cost‐efficiency disk failure prediction via threshold‐moving
    DFPE: Explaining Predictive Models for Disk Failure Prediction
    • 3
    • Highly Influenced
    Disk Failure Prediction in Data Centers via Online Learning
    • 29
    • PDF
    Disk failure prediction in heterogeneous environments
    • 15
    • PDF
    Robust Data Preprocessing for Machine-Learning-Based Disk Failure Prediction in Cloud Production Environments
    • 1
    • PDF
    Lifelong Disk Failure Prediction via GAN-Based Anomaly Detection
    • 1
    Minority Disk Failure Prediction Based on Transfer Learning in Large Data Centers of Heterogeneous Disk Systems
    • J. Zhang, K. Zhou, +5 authors Yin hu Wang
    • Computer Science
    • IEEE Transactions on Parallel and Distributed Systems
    • 2020
    • 1
    Transfer Learning based Failure Prediction for Minority Disks in Large Data Centers of Heterogeneous Disk Systems
    • 6
    AIOPS Prediction for Hard Drive Failures Based on Stacking Ensemble Model
    • H. Wang, Haili Zhang
    • Computer Science
    • 2020 10th Annual Computing and Communication Workshop and Conference (CCWC)
    • 2020
    • 1

    References

    SHOWING 1-10 OF 26 REFERENCES
    Proactive drive failure prediction for large scale storage systems
    • 75
    • PDF
    Hard Drive Failure Prediction Using Classification and Regression Trees
    • J. Li, Xinpu Ji, +4 authors X. Liu
    • Computer Science
    • 2014 44th Annual IEEE/IFIP International Conference on Dependable Systems and Networks
    • 2014
    • 67
    • PDF
    Predicting Disk Failures with HMM- and HSMM-Based Approaches
    • 47
    Improved disk-drive failure warnings
    • 171
    • PDF
    Bayesian approaches to failure prediction for disk drives
    • 182
    • PDF
    Machine Learning Methods for Predicting Failures in Hard Drives: A Multiple-Instance Application
    • 212
    • Highly Influential
    • PDF
    Proactive Data Migration for Improved Storage Availability in Large-Scale Data Centers
    • 26
    • PDF
    Disk failures in the real world : What does an MTTF of 1 , 000 , 000 hours mean to you ?
    • 278