Dependence of Reallocated Sectors Count on HDD Power-on Time

  • Authors

    • Iskandar N. Nasyrov
    • Ildar I. Nasyrov
    • Rustam I. Nasyrov
    • Bulat A. Khairullin
    2018-09-27
    https://doi.org/10.14419/ijet.v7i4.7.20544
  • information, information storage device, hard drive, reliability, reallocated sector, operating time.

  • The problem of SMART-data ambiguity in different models of hard disk drives of the same manufacturers is considered. This circumstance creates obstacles for the use of SMART technology when assessing and predicting the reliability of storage devices. The scientific task of the work is to study the dependence of the hard disk failure probability on the reliability parameters values for each individual storage device of any model of any manufacturer. In the course of the study, two interrelated parameters were analyzed: “5 Reallocated sectors count†and “9 Power-on hours†(the number of hours spent in the on state). As a result of the analysis, two types of dependences were revealed: drooping and dome shaped. The first means the maximum failure frequency of information storage devices immediately after commissioning, the second - after a certain period of time, actually coinciding with the warranty period for the products (two years). With the help of clustering in plane according to the coordinates of the number of reallocated sectors and the time of operation, two different reasons for the failure of the drives were discovered: due to deterioration of the disk surface and due to errors in the positioning of the read / write heads. Based on the variety of types of causes and consequences of equipment failure, the task of individual assessment of an individual data storage device reliability is proposed to be solved using several parameters simultaneously.

     

     

  • References

    1. [1] S.M.A.R.T. From Wikipedia, the free encyclopedia. URL: https://en.wikipedia.org/wiki/S.M.A.R.T. Checked on 10/03/2018.

      [2] Hard Drive Data and Stats / Backblaze. URL: https://www.backblaze.com/b2/hard-drive-test-data.html. Checked on 10/03/2018.

      [3] Beach B. Reliability Data Set For 41,000 Hard Drives Now Open Source. URL: https://www.backblaze.com/blog/hard-drive-data-feb2015/. Checked on 10/03/2018.

      [4] Nasyrov R.I., Nasyrov I.N., Timergaliev S.N. Cluster analysis of information storage devices that failed during operation in a large data center // Information technology. Automation. Updating and solving the problems of training highly qualified personnel (ITAP-2017): materials of the international scientific-practical conference on 19 May, 2017. - Naberezhnye Chelny: KFU, 2017. - Pз. 95-102. URL: https://cloud.mail.ru/public/LBcn/phxM8D1S5.

      [5] Pinheiro E., Weber W.-D., Barroso L.A. Failure Trends in a Large Disk Drive Population // The Proceedings of the 5th USENIX Conference on File and Storage Technologies (FAST’07). San Jose, California, USA, February 13-16, 2007. URL.

      [6] Nasyrov R.I., Nasyrov I.N. Applicability of SMART-parameters in the algorithm for predicting the reliability of data storage devices in large data centers // Innovations in life. - 2017. - No. 3 (22). - Pp. 133-146. URL: http://anodporirs.ru/Roman/+Innovations%20v%20life%20â„–%203(22)%20(3).pdf.

      [7] Beach B. How long do disk drives last? URL: https://www.backblaze.com/blog/how-long-do-disk-drives-last/. Checked on 10/03/2018.

      [8] Nasyrov R.I. Clustering of failed drives with respect to reliability and operating time in large data centers. IX Kama Readings: a collection of reports on the All-Russian Scientific and Practical Conference on 21 April 2017. - In 3 parts - Part 1. - Naberezhnye Chelny: CPI NCHI KFU. - 2017. - Pp. 129-132. URL: https://cloud.mail.ru/public/91y8/C2sNKESJt.

      [9] Rincón CAC, Paris J.-F., Vilalta R., Cheng AMK, Long DDE Disk failure prediction in heterogeneous environments // Proceedings of the International Symposium on Performance Assessment of Computer and Telecommunication Systems, SPECTS 2017.Seattle, WA, USA, July 9-12, 2017. URL: http://ieeexplore.ieee.org/document/8046776/.

      [10] Qian J., Skelton S., Moore J., Jiang H. P3: Priority based proactive prediction for soon-to-fail disks // Proceedings of the 10th IEEE International Conference on Networking, Architecture and Storage, NAS 2015. Boston, MA, USA, August 6-7, 2015. – 7255224. – p. 81-86. URL: http://ieeexplore.ieee.org/document/7255224/.

      [11] Botezatu M.M., Giurgiu I., Bogojeska J., Wiesmann D. Predicting disk replacement towards reliable data centers // Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD '16. San Francisco, California, USA, August 13-17, 2016. – p. 39-48. URL: https://dl.acm.org/citation.cfm?doid=2939672.2939699.

      [12] Chaves I.C., de Paula M.R.P., Leite L.G.M., Queiroz L., Pordeus J.P., Machado J.C. BaNHFaP: A Bayesian Network Based Failure Prediction Approach for Hard Disk Drives // Proceedings of the 5th Brazilian Conference on Intelligent Systems, BRACIS 2016. Recife, Pernambuco, BR, October 9-12, 2016. – 7839624. – p. 427-432. URL: http://ieeexplore.ieee.org/document/7839624/.

      [13] Gaber S., Ben-Harush O., Savir A. Predicting HDD failures from compound SMART attributes // Proceedings of the 10th ACM International Systems and Storage Conference, SYSTOR '17. Haifa, Israel, May 22-24, 2017. – Article No. 31. URL: https://dl.acm.org/citation.cfm?doid=3078468.3081875.

      [14] Gopalakrishnan P.K., Behdad S. Usage of product lifecycle data to detect hard disk drives failure factors // Proceedings of the ASME International Design Engineering Technical Conference. Cleveland, Ohio, USA, August 6–9, 2017. URL: http://proceedings.asmedigitalcollection.asme.org/proceeding.aspx?articleid=2662132.

  • Downloads

  • How to Cite

    N. Nasyrov, I., I. Nasyrov, I., I. Nasyrov, R., & A. Khairullin, B. (2018). Dependence of Reallocated Sectors Count on HDD Power-on Time. International Journal of Engineering & Technology, 7(4.7), 200-203. https://doi.org/10.14419/ijet.v7i4.7.20544