Apparatus and method for sharing predictive failure information on a computer network

• US 6,609,212 B1
• Filed: 03/09/2000
• Issued: 08/19/2003
• Est. Priority Date: 03/09/2000
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Smith, Gordon James
• Primary Examiner(s): Le, Dieu-Minh
• Application Number:
  US09/522,127
• Time in Patent Office: 1,258 Days
• Field of Search: 714/4, 714/39, 714/47, 714/48, 709/218, 709/224, 709/229, 709/213, 709/312
• US Class Current: 714/4.5
• CPC Class Codes: G06F 11/2257 using expert systems

Abstract

An apparatus and method shares predictive failure information between computer systems in a computer network. The shared predictive failure information allows a requester of a network resource to determine whether the resource will be available to perform the request based on its predictive failure information. According to a first embodiment, predictive failure information is written by each computer system on the network to a common storage that is accessible by one or more other computer systems on the network. When a computer system on the network needs a resource on another system, the requesting computer system can check the predictive failure status of the system that contains the needed resource by reading the predictive failure information in the common storage. If the predictive failure information indicates that the resource may perform the requested function, the requesting computer system issues the request to the resource. In a second embodiment, one or more network protocols for communicating between computer systems on the network are modified so that messages given in response to resource requests include the predictive failure status of the requested system. Thus, if a requester needs data from another computer system, a message returned from that system in response to the request preferably includes predictive failure status or information indicating whether or not the request can be granted. If the predictive failure status or information indicates that the request can be granted, the requester performs the operation on the requested computer system. In this manner, predictive failure information can be used in granting access to resources between computer systems on a network, which allows accesses to be prevented if the predictive failure information indicates that the resource is likely to fail before completing the request.

Claims

1. An apparatus comprising:

a plurality of computer systems coupled together via a network, wherein at least one of the computer systems includes corresponding predictive failure information;

a common storage accessible by the plurality of computer systems, the common storage containing predictive failure information for at least one of the plurality of computer systems; and

a predictive failure communication mechanism coupled to the network that accesses the predictive failure information in the common storage to determine whether a selected computer system on the network is able to service a request for a resource on the selected computer system.

2. The apparatus of claim 1 wherein the predictive failure information indicates whether at least one component on the corresponding computer system will likely fail within a specified time.

3. The apparatus of claim 1 wherein the predictive failure information includes performance information for at least one component on the corresponding computer system.

4. The apparatus of claim 1 wherein the predictive failure information indicates whether a resource on the corresponding computer system is capable of fulfilling a request from another computer system on the network.

5. The apparatus of claim 1 wherein the predictive failure information indicates a “

not tripped”

status if the predictive failure information on the corresponding computer system lies within predefined limits, and indicates a “

tripped”

status if the predictive failure information on the corresponding computer system lies outside of the predefined limits.

6. An apparatus comprising:

a plurality of computer systems coupled together via a network, wherein at least one of the computer systems includes corresponding predictive failure information; and

a predictive failure communication mechanism that communicates the predictive failure information to at least one other computer system on the network in a message in a network protocol for communication between computer systems on the network, wherein the message is sent in response to one computer system requesting a resource on a different computer system on the network.

7. The apparatus of claim 6 wherein the predictive failure information indicates whether at least one component on the corresponding computer system will likely fail within a specified time.

8. The apparatus of claim 6 wherein the predictive failure information includes performance information for at least one component on the corresponding computer system.

9. The apparatus of claim 6 wherein the predictive failure information indicates whether a resource on the corresponding computer system is capable of fulfilling a request from another computer system on the network.

10. The apparatus of claim 6 wherein the predictive failure information indicates a “

not tripped”

status if the predictive failure information on the corresponding computer system lies within predefined limits, and indicates a “

tripped”

status if the predictive failure information on the corresponding computer system lies outside of the predefined limits.

11. A computer-implemented method for sharing predictive failure information among a plurality of computer systems coupled together on a network, the method comprising the steps of:

determining predictive failure information for at least one component on a selected one of the plurality of computer systems;

the selected computer system writing the predictive failure information to a common storage; and

another of the plurality of computer systems reading the predictive failure information of the selected computer system from the common storage.

12. The method of claim 11 wherein the predictive failure information indicates whether at least one component on the corresponding computer system will likely fail within a specified time.

13. The method of claim 11 wherein the predictive failure information includes performance information for at least one component on the corresponding computer system.

14. The method of claim 11 wherein the predictive failure information indicates whether a resource on the corresponding computer system is capable of fulfilling a request from another computer system on the network.

15. The method of claim 11 wherein the predictive failure information indicates a “

not tripped”

status if the predictive failure information on the corresponding computer system lies within predefined limits, and indicates a “

tripped”

status if the predictive failure information on the corresponding computer system lies outside of the predefined limits.

16. A computer-implemented method for sharing predictive failure information among a plurality of computer systems coupled together on a network, the method comprising the steps of:

determining predictive failure information for at least one component on a first of the plurality of computer systems;

a second of the plurality of computer systems requesting a resource on the first computer system; and

in response to the second computer system requesting a resource on the first computer system, the first computer system returning a message to the second computer system, wherein the message includes predictive failure information for the first computer system.

17. The method of claim 16 wherein the predictive failure information indicates whether at least one component on the corresponding computer system will likely fail within a specified time.

18. The method of claim 16 wherein the predictive failure information includes performance information for at least one component on the corresponding computer system.

19. The method of claim 16 wherein the predictive failure information indicates whether a resource on the corresponding computer system is capable of fulfilling a request from another computer system on the network.

20. The method of claim 16 wherein the predictive failure information indicates a “

not tripped”

status if the predictive failure information on the corresponding computer system lies within predefined limits, and indicates a “

tripped”

status if the predictive failure information on the corresponding computer system lies outside of the predefined limits.

21. A program product comprising:

(A) a predictive failure communication mechanism that accesses predictive failure information for a plurality of networked computer systems in a common storage accessible by the plurality of computer systems to determine whether a selected computer system on the network is able to service a request for a resource on the selected computer system; and

(B) computer-readable signal bearing media bearing the predictive failure communication mechanism.

22. The program product of claim 21 wherein the signal bearing media comprises recordable media.

23. The program product of claim 21 wherein the signal bearing media comprises transmission media.

24. The program product of claim 21 wherein the predictive failure information indicates whether at least one component on the corresponding computer system will likely fail within a specified time.

25. The program product of claim 21 wherein the predictive failure information includes performance information for at least one component on the corresponding computer system.

26. The program product of claim 21 wherein the predictive failure information indicates whether a resource on the corresponding computer system is capable of fulfilling a request from another computer system on the network.

27. The program product of claim 21 wherein the predictive failure information indicates a “

not tripped”

status if the predictive failure information on the corresponding computer system lies within predefined limits, and indicates a “

tripped”

status if the predictive failure information on the corresponding computer system lies outside of the predefined limits.

28. A program product comprising:

(A) a predictive failure communication mechanism that communicates predictive failure information for a first computer system on a network to at least one other computer system on the network in a message in a network protocol for communication between computer systems on the network, wherein the message is sent in response to one computer system requesting a resource on the first computer system; and

(B) computer-readable signal bearing media bearing the predictive failure communication mechanism.

29. The program product of claim 28 wherein the signal bearing media comprises recordable media.

30. The program product of claim 28 wherein the signal bearing media comprises transmission media.

31. The program product of claim 28 wherein the predictive failure information indicates whether at least one component on the corresponding computer system will likely fail within a specified time.

32. The program product of claim 28 wherein the predictive failure information includes performance information for at least one component on the corresponding computer system.

33. The program product of claim 28 wherein the predictive failure information indicates whether a resource on the corresponding computer system is capable of fulfilling a request from another computer system on the network.

34. The program product of claim 28 wherein the predictive failure information indicates a “

not tripped”

status if the predictive failure information on the corresponding computer system lies within predefined limits, and indicates a “

tripped”

status if the predictive failure information on the corresponding computer system lies outside of the predefined limits.