System model determination for failure detection and isolation, in particular in computer systems

• US 6,587,960 B1
• Filed: 01/11/2000
• Issued: 07/01/2003
• Est. Priority Date: 01/11/2000
• Status: Expired due to Fees

• Current Assignee: Agilent Technologies, Inc.
• Original Assignee: Agilent Technologies, Inc.
• Inventors: Barford, Lee Alton, Zurhorst, Christian
• Primary Examiner(s): Iqbal, Nadeem
• Application Number:
  US09/481,077
• Time in Patent Office: 1,267 Days
• Field of Search: 714/8, 714/25, 714/26, 714/33, 714/38, 714/37, 703/17, 703/18
• US Class Current: 714/6.13
• CPC Class Codes: G06F 11/2257 using expert systems

Abstract

Disclosed is a method for determining a system model describing a relation between applicable tests and components of a system under test (SUT). The system model is applicable in conjunction with actual test results for determining at least one fault candidate representing a specific component of the SUT likely to have caused a fault of the SUT. The method system model of the SUT is determined by determining configuration information describing the actual set of components—or parts thereof—in the SUT, and combining the configuration information with static information, for dragging from the static information the information relating to the SUT. The static information describes general information about the SUT, and/or possible components of the SUT, and/or tests applicable for the SUT, and/or relations therebetween.

Claims

1. A method for determining a system model describing a relation between applicable tests and components of a system under test (SUT), whereby the system model is applicable in conjunction with actual test results for determining at least one fault candidate representing a specific component of the SUT likely to have caused a fault of the SUT, preferably provided with a certain probability that the component has caused the failure;

the method comprising the steps of;

(a) determining configuration information describing the actual set of components, or parts thereof, in the SUT, and

(b) determining the system model of the SUT by combining the configuration information with static information, for extracting from the static information the information relating to the SUT, whereby the static information describes general information about the SUT, and/or possible components of the SUT, and/or tests applicable for the SUT, and/or relations therebetween.

2. The method of claim 1, wherein steps (a) and/or (b) are iteratively repeated, preferably until no further possible components of the SUT and/or tests applicable for the SUT are determined for the system model.

3. The method of claim 1, wherein step (b) comprises a step of:

(b1) determining a set of component instances for the components of the configuration information, each component instance comprising a device identifier for uniquely identifying the component and a system address of the component in the SUT.

4. The method of claim 3, wherein step (b) further comprises a step of:

(b2) combining one or more of the component instances with the static information for determining further components present in the SUT which might not yet have been identified.

5. The method of claim 3, wherein step (b) further comprises the steps of:

(b3) combining one or more of the component instances with the static information for determining one or more tests covering at least one of the component instances, and

(b4) combining one or more of the component instances with the static information for determining for each determined test a coverage that the test has on each of those component instances.

6. The method according to claim 5, wherein step (b) further comprises a step of:

(b5) determining one or more shared functions for the determined one or more tests, whereby each shared function denotes that a specific component or component instance is tested in plural tests in the same way.

7. The method according to claim 3, wherein step (b) further comprises a step of:

(b6) combining one or more of the component instances with the static information for determining a failure probability for each one of the one or more of the component instances, whereby each failure probability represents the probability that a specific component or component instance will fail.

8. A system for determining a system model describing a relation between applicable tests and components of a system under test (SUT), whereby the system model is applicable in conjunction with actual test results for determining at least one fault candidate representing a specific component of the SUT likely to have caused a fault of the SUT, preferably provided with a certain probability that the component has caused the failure, the system comprising:

static information describing general information about the SUT, and/or possible components of the SUT, and/or tests applicable for the SUT, and/or relations therebetween, and

means for determining the system model of the SUT by combining the configuration information, describing the actual set of components—

or parts thereof—

in the SUT, with the static information, for dragging from the static information the information relating to the SUT.

9. The system of claim 8, wherein the static information comprises:

a static SUT database comprising general information about the general system type of the SUT, independent of the specific setup of the actual SUT, and/or

a static component database comprising general information about possible components applicable in the SUT, independent of the specific component setup of the actual SUT, and/or

a static test database comprising general information about possible tests executable on or applicable in the SUT, independent of the specific tests actually executed/applied for the actual SUT.