Hard disk failure prediction method for cloud computing platform

Abstract

The invention discloses a hard disk failure prediction method for a cloud computing platform. The hard disk failure predication method comprises the following steps: marking SMART log data of a hard disk as a normal hard disk sample and a faulted hard disk sample according to a hard disk maintenance record in a prediction time window; then, dividing the denoised normal hard disk sample into k non-intersected subsets by adopting a K-means clustering algorithm; combining the k non-intersected subsets with the faulted hard disk sample respectively; generating k groups of balance training sets according to an SMOTE (Synthetic Minority Oversampling Technique) so as to obtain k support vector machine classifiers for predicting the faulted hard disk. In the prediction stage, test sets can be clustered by using a DBSCAN (Density-based Spatial Clustering Of Applications With Noise), a sample in a clustered cluster is predicted as the normal hard disk sample, a noise sample is predicted by each classifier obtained by training, and further a final prediction result is obtained by voting. According to the method disclosed by the invention, hard disk fault prediction is carried out by using the SMART data of the hard disk, and relatively high fault recall ratio and overall performance can be obtained.

Description

technical field [0001] The invention relates to a hard disk failure prediction method of a cloud computing platform, which belongs to the field of computer data mining, and specifically relates to a hard disk failure prediction algorithm. Background technique [0002] Hard disk failure prediction can ensure data security, improve operation and maintenance efficiency, and control storage costs. This technology involves technologies in many fields such as cloud computing, data mining, hard disk SMART technology, fault prediction technology, and extremely unbalanced data classification technology. Hard disk failure prediction mainly refers to relying on hard disk SMART data for failure prediction. However, as document 1: PINHEIRO E, WEBERWD, BARROSO L A.Failure trends in a large disk drive population[EB / OL].[2012-10-10].http: / / research.google.com / archive / disk_failures .pdf. Introduced, using statistical methods, 36% of the causes of failures cannot be estimated. [0003] At ...

AI Technical Summary

Problems solved by technology

Cost-sensitive learning adjusts the penalty parameters according to the situation. In unbalanced classification, setting a larger penalty parameter for positive class misclassification can improve the classification effect of the classifier on the positive class. The effect of this type of method depends on the set parameters; support Compared with other classification methods, the vector machine is less sensitive to data imbalance, as in literature 8: Japkowicz N, Stephen S. The class imbalance problem: A systematic study [J]. Intelligent data analysis, 2002, In 6(5):429-449., Japkowicz et al. compared the impact of data imbalance on different classification methods, including decision tree C4.5, BP neural network and support vector machine, etc., and the results showed that support vector machine It is relatively insensitive to data imbalance, so on this issue, many methods based on support vector machines have appeared; the combination method is to combine several classifiers to improve the classification effect, and the combination method needs to combine multiple classifiers. Compromise the differences and biases of different methods, and it is easy to cause the problem of over-learning

Images