Systems and methods for detecting machine faults in network using acoustic monitoring

• US 8,607,093 B2
• Filed: 08/31/2009
• Issued: 12/10/2013
• Est. Priority Date: 08/31/2009
• Status: Active Grant

• Current Assignee: Red Hat, Inc. (International Business Machines Corporation)
• Original Assignee: Red Hat, Inc. (International Business Machines Corporation)
• Inventors: DeHaan, Michael Paul, Likins, Adrian Karstan, Vidal, Seth Kelby
• Primary Examiner(s): Patel, Kamini
• Assistant Examiner(s): BRYAN, JASON B
• Application Number:
  US12/551,418
• Publication Number:
  US 20110055669A1
• Time in Patent Office: 1,562 Days
• Field of Search: None
• US Class Current: 714/25
• CPC Class Codes: G06F 11/0709 in a distributed system consisting of a plurality of standalone computer nodes, e.g. clusters, client-server systems
• CPC Class Codes: G06F 11/0751 Error or fault detection not based on redundancy power supply failures G06F1/30; network fault management H04L41/06
• 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 11/3072 where the reporting involves data filtering, e.g. pattern matching, time or event triggered, adaptive or policy-based reporting
• CPC Class Codes: G06F 11/3089 Monitoring arrangements determined by the means or processing involved in sensing the monitored data, e.g. interfaces, connectors, sensors, probes, agents software debugging using additional hardware using a specific debug interface G06F11/3656; performance evaluation by tracing or monitoring G06F11/3466
• CPC Class Codes: H04L 41/147 for predicting network behaviour
• CPC Class Codes: H04L 41/16 using machine learning or artificial intelligence

Abstract

Embodiments relate systems and methods for detecting machine faults in a network using acoustic monitoring. In embodiments, one or more servers, clients, or other machines in a managed network can have a microphone or other acoustic sensor integrated into motherboard or other hardware. The sensor can sample acoustic signals from inside or near the machine, and can digitize that data. The resulting set of acoustic data can be transmitted to a management server or other destination for analysis of the operating sounds related to that machine. For instance, the acoustic data can be analyzed to detect indications of a failed or failing hard drive, for instance by detecting spindle whine or head movement noises, or a failed or failing power supply based on other sounds. The management server can respond to potential fault events for instance by issuing configuration commands, such as instructions to power down the malfunctioning component.

Claims

1. A method comprising:

receiving acoustic data from a machine in a managed network;

performing a spectral analysis of the acoustic data;

analyzing, by a processor, the spectral analysis in view of a plurality of event signatures stored in a recognition file, wherein each of the plurality of event signatures comprise a failure event and a plurality of conditions in the managed network, wherein the plurality of conditions comprise one of an increase or decrease in ambient sound in the managed network over a defined period;

detecting, by the processor, the failure event and one of the plurality of conditions in one of the plurality of event signatures in view of the analyzing; and

generating a configuration command to the machine in view of the detecting.

2. The method of claim 1, wherein the acoustic data comprises acoustic signals captured via acoustic sensor in the machine.

3. The method of claim 2, further comprising performing an analog to digital conversion of the acoustic signals to generate a digital representation of the acoustic signals.

4. The method of claim 3, further comprising performing the spectral analysis of the digital representation of the acoustic signals.

5. The method of claim 1, wherein the plurality of event signatures comprise one of an acoustic signature of a malfunctioning disk, an acoustic signature of a malfunctioning power supply, or an acoustic signature of a malfunction machine motherboard.

6. The method of claim 1, wherein the configuration command comprises at least a power control command.

7. The method of claim 1, wherein the acoustic data is transmitted in one of a continuous, periodic and irregular interval to the processor.

8. The method of claim 1, wherein the plurality of conditions comprise one of a whining sound from a damaged hard disk, a buzzing sound from a damaged power transformer, a thrashing sound from a hard disk excessively caching or searching data or a low background sound indicating a power or motherboard failure.

9. The method of claim 1, wherein the plurality of event signatures stored in the recognition file are organized as spectral signatures obtained via fast Fourier analysis of one of sampled damaged disks or damaged power supplies.

10. The method of claim 1, wherein the failure event is one of a potential failure event or an actual failure event.

11. The method of claim 10, wherein the configuration command comprise one of power down, park, place into sleep mode, or de-activate the machine that is identified as being in one of the potential or actual failure event.

12. A system, comprising:

a memory;

a processor coupled to the memory, wherein the processor is configured to;

receive acoustic data captured from a machine;

perform a spectral analysis of the acoustic data;

analyze the spectral analysis in view of a plurality of event signatures stored in a recognition file, wherein each of the plurality of event signatures comprise a failure event and a plurality of conditions in the managed network, wherein the plurality of conditions comprise one of an increase or decrease in ambient sound in the managed network over a defined period;

detect the failure event and one of the plurality of conditions in one of the plurality of event signatures in view of the analyzing; and

generate a configuration command to the machine in view of the detecting.

13. The system of claim 12, wherein the acoustic data comprises acoustic signals captured via acoustic sensor in the machine.

14. The system of claim 13, wherein the acoustic signals comprise a digital representation of the acoustic signals.

15. The system of claim 14, wherein the processor is further configured to perform a the spectral analysis of the digital representation of the acoustic signals.

16. The system of claim 12, wherein the plurality of event signatures comprise one of an acoustic signature of a malfunctioning disk, an acoustic signature of a malfunctioning power supply, or an acoustic signature of a malfunction machine motherboard.

17. The system of claim 12, wherein the configuration command comprises at least a power control command.

18. The system of claim 12, wherein the processor is arranged to be instantiated from a cloud-based network.

19. The system of claim 12, wherein the plurality of conditions comprise one of a whining sound from a damaged hard disk, a buzzing sound from a damaged power transformer, a thrashing sound from a hard disk excessively caching or searching data or a low background sound indicating a power or motherboard failure.

20. The system of claim 12, wherein the plurality of event signatures stored in the recognition file are organized as spectral signatures obtained via fast Fourier analysis of one of sampled damaged disks or damaged power supplies.

21. The system of claim 12, wherein the failure event is one of a potential failure event or an actual failure event.

22. The system of claim 21, wherein the configuration command comprise one of power down, park, place into sleep mode, or de-activate the machine that is identified as being in one of the potential or actual failure event.

23. A non-transitory machine-readable storage medium including data that, when accessed by a processor, cause the processor to perform operations comprising:

receiving acoustic data from a machine in a managed network;

performing a spectral analysis of the acoustic data;

analyzing, by the processor, the spectral analysis in view of a plurality of event signatures stored in a recognition file, wherein each of the plurality of event signatures comprise a failure event and a plurality of conditions in the managed network, wherein the plurality of conditions comprise one of an increase or decrease in ambient sound in the managed network over a defined period;

detecting, by the processor, the failure event and one of the plurality of conditions in one of the plurality of event signatures in view of the analyzing; and

generating a configuration command to the machine in view of the detecting.

24. The non-transitory machine-readable storage medium of claim 23, wherein the acoustic data comprises acoustic signals captured via acoustic sensor in the machine.

25. The non-transitory machine-readable storage medium of claim 24 having further data therein, which accessed by the processor, cause the processor to perform a further set of operations, further comprising:

performing an analog to digital conversion of the acoustic signals to generate a digital representation of the acoustic signals.

26. The non-transitory machine-readable storage medium of claim 25 having further data therein, which accessed by the processor, cause the processor to perform a further set of operations, further comprising:

performing the spectral analysis of the digital representation of the acoustic signals.

27. The non-transitory machine-readable storage medium of claim 23, wherein the plurality of event signatures comprise one of an acoustic signature of a malfunctioning disk, an acoustic signature of a malfunctioning power supply, or an acoustic signature of a malfunction machine motherboard.

28. The non-transitory machine-readable storage medium of claim 23, wherein the configuration command comprises at least a power control command.

29. The non-transitory machine-readable storage medium of claim 23, wherein the acoustic data is transmitted in one of a continuous, periodic and irregular interval to the processor.

30. The non-transitory machine-readable storage medium of claim 23, wherein the plurality of conditions comprise one of a whining sound from a damaged hard disk, a buzzing sound from a damaged power transformer, a thrashing sound from a hard disk excessively caching or searching data or a low background sound indicating a power or motherboard failure.

31. The non-transitory machine-readable storage medium of claim 23, wherein the plurality of event signatures stored in the recognition file are organized as spectral signatures obtained via fast Fourier analysis of one of sampled damaged disks or damaged power supplies.

32. The non-transitory machine-readable storage medium of claim 23, wherein the failure event is one of a potential failure event or an actual failure event.

33. The non-transitory machine-readable storage medium of claim 32, wherein the configuration command comprise one of power down, park, place into sleep mode, or de-activate the machine that is identified as being in one of the potential or actual failure event.