Detecting a failure condition in a system using three-dimensional telemetric impulsional response surfaces

• US 20080120064A1
• Filed: 10/26/2006
• Published: 05/22/2008
• Est. Priority Date: 10/26/2006
• Status: Active Grant

• Current Assignee: Oracle America, Inc. (Oracle Corporation)
• Original Assignee: Sun Microsystems Incorporated (Oracle Corporation)
• Inventors: Gross, Kenny C., Urmanov, Aleksey M., Bougaev, Anton A.
• Granted Patent: US 7,548,820 B2
• Time in Patent Office: Days
• US Class Current: 702/182
• CPC Class Codes: G06F 11/008 Reliability or availability analysis

Abstract

One embodiment of the present invention provides a system that facilitates high-sensitivity detection of an anomaly in telemetry data from an electronic system using a telemetric impulsional response fingerprint of the telemetry data. During operation, the system applies a sudden impulse step change to one or more operational parameters of the electronic system during operation. Next, the system generates a three-dimensional (3D) telemetric impulsional response fingerprint (TIRF) surface from a dynamic response in the telemetry data to the sudden impulse step change. The system then determines from the 3D TIRF surface whether the telemetry data contains an anomaly.

Claims

1. A method for facilitating high-sensitivity detection of an anomaly in telemetry data from an electronic system using a telemetric impulsional response fingerprint of the telemetry data, the method comprising:

applying a sudden impulse step change to one or more operational parameters of the electronic system during operation;

generating a three-dimensional (3D) telemetric impulsional response fingerprint (TIRF) surface from a dynamic response in the telemetry data to the sudden impulse step change; and

determining from the 3D TIRF surface whether the telemetry data contains an anomaly.

2. The method of claim 1, wherein determining from the 3D TIRF surface whether the telemetry data contains an anomaly involves:

generating a reference 3D TIRF surface for the 3D TIRF surface;

computing a residual response surface between the 3D TIRF surface and the reference TIRF surface; and

determining from the residual response surface whether the telemetry data contains an anomaly.

3. The method of claim 2, wherein generating the reference 3D TIRF surface for the 3D TIRF surface involves:

receiving a certified electronic system of the same type as the electronic system, wherein the certified electronic system is guaranteed to operate normally;

applying the same sudden impulse step change to the same one or more operational parameters of the certified electronic system; and

generating the reference 3D TIRF surface to the sudden impulse step change from a dynamic response in the corresponding telemetry data from the certified electronic system.

4. The method of claim 2, wherein computing the residual response surface involves subtracting the reference 3D TIRF surface from the 3D TIRF surface.

5. The method of claim 1, wherein generating the 3D TIRF surface involves:

producing a two-dimensional (2D) time-domain TIRF from the dynamic response in the telemetry data to the sudden impulse step change; and

computing a complex transfer function of the 2D time-domain TIRF, wherein the complex transfer function represents the 3D TIRF surface in a complex plane.

6. The method of claim 2, wherein determining from the residual response surface whether the telemetry data contains an anomaly involves applying a 2D Sequential Probability Ratio Test (SPRT) to the residual response surface.

7. The method of claim 6, wherein applying the 2D SPRT to the residual response surface involves performing two parallel one-dimensional (1D) SPRT tests along each direction of the residual response surface.

8. The method of claim 5, wherein the complex transfer function can be obtained by using:

a z-transform;

or

a Fourier-transform.

9. A computer-readable storage medium storing instructions that when executed by a computer cause the computer to perform a method for facilitating high-sensitivity detection of an anomaly in telemetry data from an electronic system using a telemetric impulsional response fingerprint of the telemetry data, the method comprising:

applying a sudden impulse step change to one or more operational parameters of the electronic system during operation;

generating a three-dimensional (3D) telemetric impulsional response fingerprint (TIRF) surface from a dynamic response in the telemetry data to the sudden impulse step change; and

determining from the 3D TIRF surface whether the telemetry data contains an anomaly.

10. The computer-readable storage medium of claim 9, wherein determining from the 3D TIRF surface whether the telemetry data contains an anomaly involves:

generating a reference 3D TIRF surface for the 3D TIRF surface;

computing a residual response surface between the 3D TIRF surface and the reference TIRF surface; and

determining from the residual response surface whether the telemetry data contains an anomaly.

11. The computer-readable storage medium of claim 10, wherein generating the reference 3D TIRF surface for the 3D TIRF surface involves:

receiving a certified electronic system of the same type as the electronic system, wherein the certified electronic system is guaranteed to operate normally;

applying the same sudden impulse step change to the same one or more operational parameters of the certified electronic system; and

generating the reference 3D TIRF surface to the sudden impulse step change from a dynamic response in the corresponding telemetry data from the certified electronic system.

12. The computer-readable storage medium of claim 10, wherein computing the residual response surface involves subtracting the reference 3D TIRF surface from the 3D TIRF surface.

13. The computer-readable storage medium of claim 9, wherein generating the 3D TIRF surface involves:

producing a two-dimensional (2D) time-domain TIRF from the dynamic response in the telemetry data to the sudden impulse step change; and

computing a complex transfer function of the 2D time-domain TIRF, wherein the complex transfer function represents the 3D TIRF surface in a complex plane.

14. The computer-readable storage medium of claim 10, wherein determining from the residual response surface whether the telemetry data contains an anomaly involves applying a 2D Sequential Probability Ratio Test (SPRT) to the residual response surface.

15. The computer-readable storage medium of claim 14, wherein applying the 2D SPRT to the residual response surface involves performing two parallel one-dimensional (1D) SPRT tests along each direction of the residual response surface.

16. The computer-readable storage medium of claim 13, wherein the complex transfer function can be obtained by using:

a z-transform;

or

a Fourier-transform.

17. An apparatus that facilitates high-sensitivity detection of an anomaly in telemetry data from an electronic system using a telemetric impulsional response fingerprint of the telemetry data, comprising:

an excitation mechanism configured to apply a sudden impulse step change to one or more operational parameters of the electronic system during operation;

a generation mechanism configured to generate a three-dimensional (3D) telemetric impulsional response fingerprint (TIRF) surface from a dynamic response in the telemetry data to the sudden impulse step change; and

a determination mechanism configured to determine from the 3D TIRF surface whether the telemetry data contains an anomaly.

18. The apparatus of claim 17, wherein the determination mechanism is configured to:

generate a reference 3D TIRF surface for the 3D TIRF surface;

compute a residual response surface between the 3D TIRF surface and the reference TIRF surface; and

to

determine from the residual response surface whether the telemetry data contains an anomaly.

19. The apparatus of claim 18, wherein the determination mechanism is further configured to:

receive a certified electronic system of the same type as the electronic system, wherein the certified electronic system is guaranteed to operate normally;

apply the same sudden impulse step change to the same one or more operational parameters of the certified electronic system; and

to

generate the reference 3D TIRF surface to the sudden impulse step change from a dynamic response in the corresponding telemetry data from the certified electronic system.

20. The apparatus of claim 17, wherein the generation mechanism is configured to:

produce a two-dimensional (2D) time-domain TIRF from the dynamic response in the telemetry data to the sudden impulse step change; and

to

compute a complex transfer function of the 2D time-domain TIRF, wherein the complex transfer function represents the 3D TIRF surface in a complex plane.

21. The apparatus of claim 18, wherein the determination mechanism is further configured to apply a 2D Sequential Probability Ratio Test (SPRT) to the residual response surface.