Detecting constrained motion of a component in a computer system

• US 8,489,343 B2
• Filed: 09/12/2008
• Issued: 07/16/2013
• Est. Priority Date: 09/12/2008
• Status: Active Grant

• Current Assignee: Oracle America, Inc. (Oracle Corporation)
• Original Assignee: Oracle America, Inc. (Oracle Corporation)
• Inventors: Urmanov, Aleksey M., Bougaev, Anton A., McElfresh, David K.
• Primary Examiner(s): Huynh, Phuong
• Application Number:
  US12/209,745
• Publication Number:
  US 20100070210A1
• Time in Patent Office: 1,768 Days
• Field of Search: 702/56, 702/141, 702 33- 36, 702/39, 702/85, 702/108, 702/113, 702/115, 702/127, 702182-186, 73570-572, 73577-583, 736/45
• US Class Current: 702/56
• CPC Class Codes: G06F 11/3034 where the computing system component is a storage system, e.g. DASD based or network based digital input from or digital output to record carriers G06F3/06; digital recording or reproducing G11B20/18; for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS], H04L67/1097
• CPC Class Codes: G06F 11/3058 Monitoring arrangements for monitoring environmental properties or parameters of the computing system or of the computing system component, e.g. monitoring of power, currents, temperature, humidity, position, vibrations thermal management in cooling arrangements of a computing system G06F1/206
• CPC Class Codes: G06F 11/3409 for performance assessment

Abstract

Some embodiments of the present invention provide a system that detects the presence of constrained motion in one or more components in a computer system. First, a vibrational spectrum of the computer system is monitored while the computer system operates. Then, the vibrational spectrum is analyzed using a pattern-recognition model to detect constrained motion in one or more components in the computer system, wherein the pattern-recognition model classifies the vibrational spectrum as indicating constrained motion or as not indicating constrained motion.

Claims

1. A method for detecting the presence of constrained motion in one or more components in a computer system, the method comprising:

monitoring a vibrational spectrum of the computer system while the computer system operates;

analyzing the vibrational spectrum using a pattern-recognition model to detect constrained motion in one or more components in the computer system, wherein constrained motion occurs when at least one mounting fastener for a component is loose, so that the component impacts one or more fasteners, mounts, or other components in the computer system when vibrating, and wherein the pattern-recognition model classifies the vibrational spectrum as indicating constrained motion or as not indicating constrained motion;

monitoring one or more performance parameters in the computer system;

upon classifying the vibrational spectrum as indicating constrained motion, using the vibrational spectrum and the monitored performance parameters to identify which of the components in the computer system is undergoing the constrained motion; and

sending a signal to alter a speed of a fan for the computer system until the constrained motion for at least one of the identified components is reduced or eliminated.

2. The method of claim 1, wherein:

the constrained motion includes rattling; and

the one or more components include at least one of;

a disk drive;

a disk carrier;

a disk cage;

a fan;

a fan tray; and

a fan mount.

3. The method of claim 1, wherein prior to monitoring the vibrational spectrum, the method further comprises:

training the pattern-recognition model using a first training vibrational spectrum representing constrained motion of one or more components in a test computer system, and a second training vibrational spectrum representing an absence of constrained motion of one or more components in a test computer system.

4. The method of claim 1, wherein analyzing the vibrational spectrum includes extracting features from the vibrational spectrum and analyzing the extracted features using the pattern-recognition model.

5. The method of claim 4, wherein extracting the features from the vibrational spectrum includes generating a feature vector based on the vibrational spectrum.

6. The method of claim 1, wherein the pattern-recognition model includes one or more of:

a quadratic classifier;

a k-nearest neighbor classifier; and

an R-cloud classifier.

7. The method of claim 1, wherein analyzing the vibrational spectrum includes:

transmitting information related to the vibrational spectrum over a network to a vibrational-analysis computer system; and

analyzing the vibrational spectrum using the vibrational-analysis computer system.

8. The method of claim 1, further including generating a request for a service action based on the classification of the vibrational spectrum as indicating constrained motion or as not indicating constrained motion.

9. The method of claim 1,

wherein monitoring the vibrational spectrum includes monitoring a set of vibrational spectra of the computer system while the computer system operates;

wherein analyzing the vibrational spectrum includes analyzing the vibrational spectra using a pattern-recognition model to detect constrained motion in one or more components in a computer system, wherein the pattern-recognition model classifies each vibrational spectrum in the vibrational spectra as indicating constrained motion or as not indicating constrained motion; and

wherein the method further comprises generating a request for a service action based on the classification of a predetermined number of the vibrational spectra in the set of vibrational spectra as indicating constrained motion.

10. The method of claim 1, wherein using the vibrational spectrum and the monitored performance parameters to identify the component undergoing the constrained motion comprises identifying the component based on a relationship between a vibrational spectrum for the component that the pattern-recognition model classified as indicating constrained motion and speeds for each component in a set of components in the computer system at the time the vibrational spectrum was monitored, wherein the relationship comprises a presence of harmonics of a frequency for the component in the vibrational spectrum, and wherein the set of components comprises at least one of a fan and a disk drive.

11. The method of claim 1, wherein the at least one of the identified components do not include the fan.

12. The method of claim 11, further comprising:

continuing to send the signal until the vibrational spectrum no longer indicates constrained motion for the at least one of the identified components.

13. A non-transitory computer-readable storage medium storing instructions that when executed by a computer cause the computer to perform a method for detecting the presence of constrained motion in one or more components in a computer system, the method comprising:

monitoring a vibrational spectrum of the computer system while the computer system operates;

analyzing the vibrational spectrum using a pattern-recognition model to detect constrained motion in one or more components in the computer system, wherein constrained motion occurs when at least one mounting fastener for a component is loose, so that the component impacts one or more fasteners, mounts, or other components in the computer system when vibrating, and wherein the pattern-recognition model classifies the vibrational spectrum as indicating constrained motion or as not indicating constrained motion;

monitoring one or more performance parameters in the computer system;

upon classifying the vibrational spectrum as indicating constrained motion, using the vibrational spectrum and the monitored performance parameters to identify which of the components in the computer system is undergoing the constrained motion; and

sending a signal to alter a speed of a fan for the computer system until the constrained motion for at least one of the identified components is reduced or eliminated.

14. The computer-readable storage medium of claim 13, wherein:

the constrained motion includes rattling; and

the one or more components include at least one of;

a disk drive;

a disk carrier;

a disk cage;

a fan;

a fan tray; and

a fan mount.

15. The computer-readable storage medium of claim 13, wherein prior to monitoring the vibrational spectrum, the method further comprises:

training the pattern-recognition model using a first training vibrational spectrum representing constrained motion of one or more components in a test computer system, and a second training vibrational spectrum representing an absence of constrained motion of one or more components in a test computer system.

16. The computer-readable storage medium of claim 13, wherein analyzing the vibrational spectrum includes extracting features from the vibrational spectrum and analyzing the extracted features using the pattern-recognition model.

17. The computer-readable storage medium of claim 16, wherein extracting the features from the vibrational spectrum includes generating a feature vector based on the vibrational spectrum.

18. The computer-readable storage medium of claim 13, wherein the pattern-recognition model includes one or more of:

a quadratic classifier;

a k-nearest neighbor classifier; and

an R-cloud classifier.

19. The computer-readable storage medium of claim 13,

wherein monitoring the vibrational spectrum includes monitoring a set of vibrational spectra of the computer system while the computer system operates;

wherein analyzing the vibrational spectrum includes analyzing the vibrational spectra using a pattern-recognition model to detect constrained motion in one or more components in a computer system, wherein the pattern-recognition model classifies each vibrational spectrum in the vibrational spectra as indicating constrained motion or as not indicating constrained motion; and

wherein the method further comprises generating a request for a service action based on the classification of a predetermined number of the vibrational spectra in the set of vibrational spectra as indicating constrained motion.

20. An apparatus that detects the presence of constrained motion in one or more components in a computer system, the apparatus comprising:

a monitoring mechanism configured to monitor a vibrational spectrum of the computer system while the computer system operates and one or more performance parameters in the computer system;

an analyzing mechanism configured to analyze the vibrational spectrum using a pattern-recognition model to detect constrained motion in one or more components in the computer system, wherein constrained motion occurs when at least one mounting fastener for a component is loose, so that the component impacts one or more fasteners, mounts, or other components in the computer system when vibrating, and wherein the pattern-recognition model classifies the vibrational spectrum as indicating constrained motion or as not indicating constrained motion;

a mechanism configured to, upon the pattern-recognition model classifying the vibrational spectrum as indicating constrained motion, use the vibrational spectrum and the monitored performance parameters to identify which of the components in the computer system is undergoing the constrained motion; and

a mechanism configured to send a signal to alter a speed of a fan for the computer system until the pattern-recognition model indicates that the constrained motion for at least one of the identified components is reduced or eliminated.