Review of techniques for predicting hard drive failure with SMART attributes Online publication date: Tue, 03-Jul-2018
by Marco Garcia; Vladimir Ivanov; Anastasia Kozar; Stanislav Litvinov; Alexey Reznik; Vitaly Romanov; Giancarlo Succi
International Journal of Machine Intelligence and Sensory Signal Processing (IJMISSP), Vol. 2, No. 2, 2018
Abstract: Hard drive failure prediction is still a relevant problem today. A number of statistical and machine learning techniques were proposed to improve failure forecasting accuracy after SMART was introduced. SMART is a diagnostics tool that aims at providing forehand failure warnings. Failure prediction methods can be viewed as a part of reliability analysis - the field that was studied intensively for decades. However, in some situations available techniques cannot be applied due to a simple reason - information at hand is not always sufficient for reliable prediction. SMART's goal is to provide meaningful information that can signify problems with the health condition of a hard drive and failure prediction techniques can leverage this data to provide timely and reliable warnings. To find the best failure forecasting algorithm and evaluate the possibility of its widespread deployment, we review existing datasets with SMART attributes, methods for feature selection for hard drive failure prediction.
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Machine Intelligence and Sensory Signal Processing (IJMISSP):
Login with your Inderscience username and password:
Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.
If you still need assistance, please email subs@inderscience.com