Wavelet based hard disk analysis

• US 8,094,015 B2
• Filed: 01/22/2009
• Issued: 01/10/2012
• Est. Priority Date: 01/22/2009
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Borghetti, Stefano, Della Corte, Gianluca, Gianfagna, Leonida, Sgro′, Antonio
• Primary Examiner(s): Lu, Shirley
• Application Number:
  US12/357,913
• Publication Number:
  US 20100182158A1
• Time in Patent Office: 1,083 Days
• Field of Search: 340/540, 702/30, 702/183
• US Class Current: 340/540
• CPC Class Codes: G11B 2220/2516 Hard disks
• CPC Class Codes: G11B 27/36 Monitoring, i.e. supervising the progress of recording or reproducing for digital recording G11B20/00 and s.gr.; for monitoring, testing or measuring of TV recorders of the type covered by H04N5/76 and subgroups, see H04N17/06

Abstract

Monitoring systems with predictive failure analysis technology (e.g., Self Monitoring Analysis and Reporting Technology (S.M.A.R.T)) operate on the basis of correlations between the hard disk'"'"'s mechanical and electronic characteristics. Using wavelet transforms to analyze electrical signals from the hard disks can help better determine changes in values and trends in various hard disk characteristics. Using wavelet transforms allows for thorough and deeper analysis of the electrical signals from the hard disks. Wavelet analysis can be performed on the electrical signal to extract time and frequency information associated with the low and high frequency components. Analyzing the extracted components of the electrical signal can indicate a presence and position of one or more mechanical defects on the hard disk.

Claims

1. A method comprising:

receiving a signal from one or more sensors on a hard disk, wherein the signal from the one or more sensors indicates a variation in a characteristic of the hard disk;

determining a wavelet decomposition level based on the received signal from the one or more sensors on the hard disk;

performing wavelet analysis on the received signal from the one or more sensors on the hard disk based, at least in part, on the determined wavelet decomposition level, wherein said performing the wavelet analysis on the received signal comprises,

determining a first approximation signal and a first detail signal from the received signal, and

performing wavelet analysis on the first approximation signal to generate a second approximation signal and a second detail signal;

determining the presence of one or more spikes in the first and the second detail signals;

determining that the received signal indicates anomalies corresponding to deformations on the hard disk based, at least in part, on the presence of the one or more spikes in the first and the second detail signals; and

generating an alert indicating one or more mechanical defects on the hard disk.

2. The method of claim 1, wherein the determining the wavelet decomposition level based on the received signal comprises determining an entropy level of the received signal.

3. The method of claim 1, wherein the determining that the received signal indicates anomalies corresponding to deformations on the hard disk further comprises:

comparing the wavelet analyzed signal with one or more pre-defined hard disk models; and

determining a discrepancy between the wavelet analyzed signal and the one or more pre-defined hard disk templates.

4. The method of claim 3, wherein the one or more pre-defined hard disk models comprise any one or more of performance values, trends, waveforms and wherein the one or more pre-defined hard disk models indicate the performance of one or more of failed and ideal hard disks.

5. The method of claim 1, wherein the received signal generated by one or more of the sensors on the hard disk indicates at least one of the hard disk'"'"'s acceleration and the hard disk'"'"'s velocity.

6. A method comprising:

receiving a signal from a hard disk, wherein the received signal indicates a variation in a characteristic of the hard disk;

performing wavelet analysis on the received signal, wherein said performing wavelet analysis on the received signal comprises,

determining a first approximation signal and a first detail signal from the received signal, and

performing wavelet analysis on the first approximation signal to generate a second approximation signal and a second detail signal;

analyzing the first detail signal and the second detail signal to determine the presence of one or more spikes in the first and the second detail signals;

determining one or more anomalies in the received signal based on said analyzing the first detail signal and the second detail signal to determine the presence of one or more spikes in the first and the second detail signals, wherein the presence of the one or more spikes in the first and the second detail signals indicates one or more anomalies in the received signal; and

generating a hard disk alert in response to the presence of the one or more anomalies in the received signal.

7. The method of claim 6, wherein said generating the hard disk alert in response to the one or more anomalies in the received signal comprises one or more of turning on an LED and generating a status message.

8. One or more machine-readable media having stored therein a program product, which when executed a set of one or more processor units causes the set of one or more processor units to perform operations that comprise:

receiving a signal that indicates a variation in a characteristic of a hard disk;

performing wavelet analysis on the received signal, wherein said operation of performing wavelet analysis on the received signal comprises,

determining a first approximation signal and a first detail signal from the received signal,

performing wavelet analysis on the first approximation signal to generate a second approximation signal and a second detail signal;

analyzing the first detail signal and the second detail signal to determine the presence of one or more spikes in the first and the second detail signals;

determining one or more anomalies in the received signal based on the presence of one or more spikes in the first and the second detail signals, wherein the presence of the one or more spikes in the first and the second detail signals indicates one or more anomalies in the received signal; and

generating a hard disk alert in response to the one or more anomalies in the received signal.

9. The machine-readable media of claim 8, wherein the operations further comprise determining a wavelet decomposition level based on the received signal, wherein said operation of performing wavelet analysis on the received signal further comprises performing wavelet analysis in accordance with the determined wavelet decomposition level.

10. The machine-readable media of claim 9, wherein said operation of determining the wavelet decomposition level based on the received signal comprises determining an entropy level of the received signal.

11. The machine-readable media of claim 8, wherein said operation of generating the hard disk alert in response to the one or more anomalies in the received signal comprises one or more of turning on an LED and generating a status message.

12. The machine-readable media of claim 8, wherein said operation of determining one or more anomalies in the received signal based on the presence of one or more spikes in the first and the second detail signals further comprises:

comparing the wavelet analyzed signal with one or more pre-defined hard disk models; and

determining a discrepancy between the wavelet analyzed signal and the one or more pre-defined hard disk templates.

13. An apparatus comprising:

a set of one or more processors;

a network interface coupled with the set of one or more processors; and

a hard disk wavelet analyzer configured to

receive a signal from a hard disk, wherein the received signal indicates a variation in a characteristic of the hard disk;

perform wavelet analysis on the received signal, wherein the hard disk wavelet analyzer configured to perform wavelet analysis on the received signal comprises the hard disk wavelet analyzer being configured to,

determine a first approximation signal and a first detail signal from the received signal,

perform wavelet analysis on the first approximation signal to generate a second approximation signal and a second detail signal;

analyze the first detail signal and the second detail signal to determine the presence of one or more spikes in the first and the second detail signals;

determine one or more anomalies in the received signal based on the presence of one or more spikes in the first and the second detail signals, wherein the presence of the one or more spikes in the first and the second detail signals indicates one or more anomalies in the received signal; and

generate a hard disk alert in response to the one or more anomalies in the received signal.

14. The apparatus of claim 13, wherein the hard disk wavelet analyzer is further configured to determine a wavelet decomposition level based on the received signal, wherein the hard disk wavelet analyzer configured to perform wavelet analysis on the received signal from the one or more sensors on the hard disk further comprises the hard disk wavelet analyzer configured to perform wavelet analysis in accordance with the determined wavelet decomposition level.

15. The apparatus of claim 13, wherein the hard disk wavelet analyzer configured to determine the wavelet decomposition level based on the received signal comprises the hard disk wavelet analyzer configured to determine an entropy level of the received signal.

16. The apparatus of claim 13, wherein the hard disk wavelet analyzer comprises one or more machine-readable media.