Method and device for extracting feature data from SMART data of hard disk

Abstract

The invention discloses a method and a device for extracting feature data from SMART self-monitoring analysis and reporting technology data of a hard disk. The method comprises the following steps: obtaining an SMART data set from a sample hard disk, wherein the SMART data set comprises Q SMART data and Q hard disk type information corresponding to the Q SAMART data respectively; normalizing the Q SMART data to generate Q normalized SMART data; correcting the Q normalized SMART data according to the Q hard disk type information respectively to generate a corrected SMART data set; generating the feature data of the hard disk according to the corrected SMART data set. According to the method of the embodiment disclosed by the invention, different hard disks can be subjected to fault prewarning test and analysis through a same fault prewarning model, so that the accuracy of the fault prewarning model is improved and the model training, test and analysis costs can be reduced.

Description

technical field [0001] The invention relates to the field of storage technology, in particular to a method and device for extracting feature data from hard disk SMART self-monitoring, analysis and reporting technical data. Background technique [0002] Since the hard disk failure can be reflected from the hard disk SMART (Self Monitoring Analysis And Reporting Technology, self-monitoring, analysis and reporting technology) data, in the hard disk failure early warning analysis, the hard disk can be analyzed according to the SMART data of the hard disk. whether there will be a failure in the At present, machine learning algorithms can be used to train a failure warning model based on a certain attribute in the SMART data, and then analyze the SMART data of the hard disk according to the failure warning model to predict whether the hard disk can work stably in the future. [0003] However, because the eigenvalue representations of different attributes in SMART data are not uni...

AI Technical Summary

Problems solved by technology

[0003] However, because the eigenvalue representations of different attributes in SMART data are not uniform and too discrete, it is difficult to predict the common impact of several different attributes on the hard disk.

Moreover, some attributes have missing eigenvalues ​​when training the model, which increases the difficulty of analyzing SMART data and makes the model prediction inaccurate

In addition, the eigenvalue calculation methods of hard disk data from different manufacturers are not uniform, which is not conducive to a unified numerical feature representation. Therefore, it is necessary to train the fault early warning model for the SMART data of each manufacturer's hard disk for fault early warning analysis, which requires multiple times. Perform model training, which greatly increases the cost of analysis

Images