ABSTRACT
Disk failure has become a major concern with the rapid expansion of storage systems in data centers. Based on SMART (Self-Monitoring, Analysis and Reporting Technology) attributes, many researchers derive disk failure prediction models using machine learning techniques. Despite the significant developments, the majority of works rely on offline training and thereby hinder their adaption to the continuous update of forthcoming data, suffering from the 'model aging' problem. We are therefore motivated to uncover the root cause -- the dynamic SMART distribution for 'model aging', aiming to resolve the performance degradation as to pave a comprehensive study in practice.
In this paper, we introduce a novel disk failure prediction model using Online Random Forests (ORFs). Our ORF-based model can automatically evolve with sequential arrival of data on-the-fly and thus is highly adaptive to the variance of SMART distribution over time. Moreover, it has favourable advantage against the offline counterparts in terms of superior prediction performance. Experiments on real-world datasets show that our ORF model converges rapidly to the offline random forests and achieves stable failure detection rates of 93-99% with low false alarm rates. Furthermore, we demonstrate the ability of our approach on maintaining stable prediction performance for the long-term usage in data centers.
- B. Schroeder and G. A. Gibson. Disk failures in the real world: What does an MTTF of 1,000,000 hours mean to you? In Proceedings of 5th USENIX Conference on File and Storage Technologies (FAST'07), pp. 1--16, Feb. 2007. Google ScholarDigital Library
- I. Manousakis, S. Sankar, G. McKnight, Thu D. Nguyen, and R. Bianchini. Environmental Conditions and Disk Reliability in Free-Cooled Datacenters. In Proceedings of the 14th USENIX Conference on File and Storage Technologies (FAST'16), pp. 53--65, Feb. 2016. Google ScholarDigital Library
- Data center downtime costs. http://www.emerson.com/en-us/News/Pages/Net-Power-Study-Data-Center.aspx.Google Scholar
- L. N. Bairavasundaram, G. R. Goodson, B. Schroeder, A. C. Arpaci-Dusseau, and R. H. Arpaci-Dusseau. An analysis of data corruption in the storage stack. In Proceedings of the 6th USENIX Conference on File and Storage Technologies (FAST'08), Feb. 2008. Google ScholarDigital Library
- E. Brewer, L. Ying, L. Greenfield, R. Cypher, and Theodore. Disks for Data Centers. Keynote talk for FAST'16, Technical report, Google, Feb. 2016.Google Scholar
- P. M. Chen, E. K. Lee, G. A. Gibson, R. H. Katz, and D. A. Patterson. RAID: High-performance, reliable secondary storage. In ACM Computing Surveys (CSUR), vol 26, no. 2, pp. 145--185, June, 1994. Google ScholarDigital Library
- S.M.A.R.T. https://en.wikipedia.org/wiki/S.M.A.R.T.Google Scholar
- G. Hamerly and C. Elkan. Bayesian approaches to failure prediction for disk drives. In Proceedings of the 18th International Conference on Machine Learning (ICML'01), pp. 202--209, Jun. 2001. Google ScholarDigital Library
- G. F. Hughes, J. F. Murray, K. Kreutz-Delgado, and C. Elkan. Improved disk-drive failure warnings. In IEEE Transactions on Reliability, vol. 51, no. 3, pp. 350--357, Sept. 2002.Google ScholarCross Ref
- J. F. Murray, G. F. Hughes, and K. Kreutz-Delgado. Machine learning methods for predicting failures in hard drives: A multiple instance application. In Journal of Machine Learning Research, vol. 6, pp. 783--816, May. 2005. Google ScholarDigital Library
- B. Zhu, G. Wang, X. Liu, D. Hu, S. Lin, and J. Ma. Proactive drive failure prediction for large scale storage systems. In Proceedings of 29th IEEE Conference on Massive Storage Systems and Technologies (MSST), pp. 1--5, May. 2013.Google ScholarCross Ref
- Y. Wang, Q. Miao, E. W. Ma, K.-L. Tsui, and M. G. Pecht. Online anomaly detection for hard disk drives based on mahalanobis distance. In IEEE Transactions on Reliability, vol. 62, no. 1, pp. 136--145, Mar. 2013.Google ScholarCross Ref
- Y. Wang, W.M. Ma, W. S. Chow, and K.-L. Tsui. A Two-Step Parametric Method for Failure Prediction in Hard Disk Drives. In IEEE Transactions on Industrial Informatics, vol. 10, no. 1, pp. 419--430, Feb. 2014.Google ScholarCross Ref
- B. Zhu, G. Wang, X. Ji, J. Li, and Y. Jia. Hard Drive Failure Prediction Using Classification and Regression Trees. In Proceedings of 44th IEEE Conference on Dependable Systems and Networks (DSN), pp. 383--394, Dec. 2014. Google ScholarDigital Library
- C. Xu, G. Wang, X. Liu, D. Guo, and T. Liu. Health Status Assessment and Failure Prediction for Hard Drives with Recurrent Neural Networks. In IEEE Transactions on Computers, vol. 65, no. 11, pp. 3502--3508, Nov. 2016. Google ScholarDigital Library
- J. Li, R. J. Stones, G. Wang, X. Liu, Z. Li, and M. Xu. Hard drive failure prediction using Decision Trees. In Reliability Engineering and System Safety, vol. 164, pp. 55--65, Mar. 2017.Google ScholarCross Ref
- J. Li, R. J. Stones, G. Wang, Z. Li, X. Liu, and X. Kang. Being accurate is not enough: New metrics for disk failure prediction. In Proceedings Symposium on Reliable Distributed Systems (SRDS), pp. 71--80, Dec. 2016.Google ScholarCross Ref
- O. Fontenla-Romero, B. Guijarro-Berdinas, D. Martinez-Rego, B. Perez-Sanchez, and D. Peteiro-Barral. Online machine learning. In Efficiency and Scalability Methods for Computational Intellect, pp. 27--54, 2013.Google ScholarCross Ref
- The backblaze hard drive data and stats. https://www.backblaze.com/b2/hard-drive-test-data.html.Google Scholar
- M. Botezatu, I. Giurgiu, J. Bogojeska, and D. Wiesmann. Predicting Disk Replacement towards Reliable Data Centers. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD), pp. 39--48, Aug. 2016. Google ScholarDigital Library
- N. C. Oza and S. J. Russell. Online bagging and boosting. In Artificial Intelligence and Statistics, pp. 105--112, 2001.Google Scholar
- A. Saffari, C. Leistner, J. Santner, M. Godec, and H. Bischof. On-line Random Forests. In Proceedings of 12th International Conference on Computer Vision Workshops (ICCV), pp. 1393--1400, 2009.Google ScholarCross Ref
- V. Agrawal, C. Bhattacharyya, T. Niranjan, and S. Susarla. Discovering rules from disk events for predicting hard drive failures. In Proceedings of International Conference on Machine Learning and Applications, pp. 782--786, Dec. 2009. Google ScholarDigital Library
- L. Breiman. Random forests. Machine Learning, vol. 45, pp. 5--32, Oct. 2001. Google ScholarDigital Library
- F. Mahdisoltani, I. Stefanovici, and B. Schroeder. Proactive error prediction to improve storage system reliability. In Proceedings of the 2017 USENIX Annual Technical Conference (USENIX ATC '17), pp. 391--402, Jul. 2017. Google ScholarDigital Library
- L. Breiman. Out-of-bag Estimation. Citeseer, 1996.Google Scholar
- E. Pinheiro, W.-D. Weber, and L. A. Barroso. Failure trends in a large disk drive population. In Proceedings of 5th USENIX Conference on File and Storage Technologies (FAST'07), pp. 17--29, Feb. 2007. Google ScholarDigital Library
- C.-C. Chang and C.-J. Lin. LIBSVM: A library for support vector machines. In ACM Transactions on Intelligent Systems and Technology, vol. 2, no. 3, pp. 27:1--27:27, Apr. 2011. Google ScholarDigital Library
- Fit binary classification tree for multiclass classification. https://cn.mathworks.com/help/stats/fitctree.html.Google Scholar
Recommendations
Transfer Learning based Failure Prediction for Minority Disks in Large Data Centers of Heterogeneous Disk Systems
ICPP '19: Proceedings of the 48th International Conference on Parallel ProcessingThe storage system in large scale data centers is typically built upon thousands or even millions of disks, where disk failures constantly happen. A disk failure could lead to serious data loss and thus system unavailability or even catastrophic ...
Multiple-Instance Learning for Hard Disk Drive Failure Prediction
ICECC '12: Proceedings of the 2012 International Conference on Electronics, Communications and ControlA hard disk drive (HDD) is a critical component in computers. The failure of HDD can result in the users' data loss and computer downtime. Both consequences cause inconveniences to the users. Therefore, detecting the impending failure of HDDs becomes a ...
Reliability and security of RAID storage systems and D2D archives using SATA disk drives
Information storage reliability and security is addressed by using personal computer disk drives in enterprise-class nearline and archival storage systems. The low cost of these serial ATA (SATA) PC drives is a tradeoff against drive reliability design ...
Comments