Methods and arrangements for ordering changes in computing systems

• US 7,496,912 B2
• Filed: 02/27/2004
• Issued: 02/24/2009
• Est. Priority Date: 02/27/2004
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Wolf, Joel L., Wu, Kun-Lung, Keller, Alexander, Hellerstein, Joseph L., Krishnan, Vijaya
• Primary Examiner(s): Zhen; Wei Y
• Assistant Examiner(s): Chen; Qing
• Application Number:
  US10/789,123
• Publication Number:
  US 20050192979A1
• Time in Patent Office: 1,824 Days
• Field of Search: 717120-123, 717168-178
• US Class Current: 717/174
• CPC Class Codes: G06F 8/61 Installation
• CPC Class Codes: H04L 41/082 the condition being updates or upgrades of network functionality
• CPC Class Codes: H04L 41/0886 Fully automatic configuration

Abstract

Methods and arrangements for automatically determining allowable sequences of changes, e.g., sequences where the order in which changes are carried out will transition a computing system from a workable state into another workable state, are disclosed.

Claims

1. A computer-implemented method of automatically determining an allowable order of changes in a distributed system, the computer-implemented method comprising the steps of:

receiving a request for change;

wherein the request for change describes a task to be done on at least one target system, and a deadline by which a change needs to be completed;

determining existing relationship descriptions between components of the distributed system, wherein the components of the distributed system are implemented on a plurality of managed resources;

transforming acquired relationships into ordered tasks that are linked by temporal ordering constraints which describe when tasks can begin in relation to one another, wherein the temporal ordering constraints are selected from the group consisting of;

Finish-to-Start, Start-to-Start, Finish-to-Finish, and Start-to-Finish;

creating an order of changes taking into account the temporal ordering constraints, wherein creating the order of changes includes determining whether the ordered changes are conflicting and flagging such conflicts and further includes an estimate of the time required to complete a change;

wherein the order of changes transitions the at least one target system from one workable state into another workable state;

building a Task Graph from the order of changes, wherein the Task Graph represents each task to be completed within an overall job;

constructing an Annotated Task Graph by assigning estimated durations to each task within the Task Graph; and

computing a makespan for the overall job represented by the Task Graph, wherein the Annotated Task Graph is returned to an administrator.

2. The computer-implemented method of claim 1, wherein the order of changes is sequential.

3. The computer-implemented method of claim 1, wherein the order of changes is concurrent.

4. The computer-implemented method of claim 1, further comprising refining an incoming request for change by breaking the incoming request down into sub-requests.

5. The computer-implemented method of claim 4, further comprising computing an allowable order of changes by interacting with the distributed system.

6. The computer-implemented method of claim 4, wherein a total change time is minimized by exploiting parallelism between change tasks.

7. The computer-implemented method of claim 1, wherein the ordered changes are partially ordered.

8. The computer-implemented method of claim 1, wherein the ordered changes are totally ordered.

9. The computer-implemented method of claim 1, wherein the creation of the order of changes further takes into account a requested change management operation.

10. The computer-implemented method of claim 1, wherein a requester identifies one or more target systems within the distributed system by name.

11. The computer-implemented method of claim 10, wherein the names of the target systems are unique physical identifiers.

12. The computer-implemented method of claim 10, wherein the names of the target systems are logical names which refer to one or more physical systems.

13. The computer-implemented method of claim 1, wherein a requester does not identify one or more target systems within the distributed system by name.

14. The computer-implemented method of claim 1, further comprising the steps of accessing and evaluating policy roles representing best practices.

15. The computer-implemented method of claim 14, wherein the best practices include updating all affected software artifacts when a software artifact is updated.

16. The computer-implemented method of claim 14, wherein the best practices include having a given set of software components installed on different systems.

17. The computer-implemented method of claim 1, wherein one or more of the order of changes are persistently stored after being created.

18. The computer-implemented method of claim 1, wherein a component is one of a service, an application, middleware, hardware, an operating system, a storage system, a network device, and a system associated with a computing environment.

19. A system for automatically determining an allowable order of changes in a distributed system, the system comprising:

a processor; and

a memory storing code accessible by the processor to;

receiving a request for change;

wherein the request for change describes a task to be done on at least one target system, and a deadline by which a change needs to be completed;

determining existing relationship descriptions between components of the distributed system, wherein the components of the distributed system are implemented on a plurality of managed resources;

transforming acquired relationships into ordered tasks that are linked by temporal ordering constraints which describe when tasks can begin in relation to one another, wherein the temporal ordering constraints are selected from the group consisting of;

Finish-to-Start, Start-to-Start, Finish-to-Finish, and Start-to-Finish;

creating an order of changes taking into account the temporal ordering constraints, wherein creating the order of changes includes determining whether the ordered changes are conflicting and flagging such conflicts and further includes an estimate of the time required to complete a change;

wherein the order of changes transitions the at least one target system from one workable state into another workable state;

building a Task Graph from the order of changes, wherein the Task Graph represents each task to be completed within an overall job;

constructing an Annotated Task Graph by assigning estimated durations to each task within the Task Graph; and

computing a makespan for the overall job represented by the Task Graph, wherein the Annotated Task Graph is returned to an administrator.

20. The system of claim 19, wherein the order of changes is sequential.

21. The system of claim 19, wherein the order of changes is concurrent.

22. The system of claim 19, further comprising an arrangement for refining an incoming request for change by breaking the incoming request down into sub-requests.

23. The system of claim 22, further comprising an arrangement for computing an allowable order of changes by interacting with the distributed system.

24. The system of claim 22, wherein a total change time is minimized by exploiting parallelism between change tasks.

25. The system of claim 19, wherein the ordered changes are partially ordered.

26. The system of claim 19, wherein the ordered changes are totally ordered.

27. The system of claim 19, wherein the creation of the order of changes further takes into account a requested change management operation.

28. The system of claim 19, wherein a requester identifies one or more target systems within the distributed system by name.

29. The system of claim 28, wherein the names of the target systems are unique physical identifiers.

30. The system of claim 28, wherein the names of the target systems are logical names which refer to one or more physical systems.

31. The system of claim 19, wherein a requester does not identify one or more target systems within the distributed system by name.

32. The system of claim 19, further comprising an arrangement for accessing and evaluating policy rules representing best practices.

33. The system of claim 32, wherein the best practices include updating all affected software artifacts when a software artifact is updated.

34. The system of claim 32, wherein the best practices include having a given set of software components installed on different systems.

35. The system of claim 19, wherein one or more of the order of changes are persistently stored after being created.

36. The system of claim 19, wherein a component is one of a service, an application, middleware, hardware, an operating system, a storage system, a network device, and a system associated with a system associated with computing environment.

37. A program storage device readable by machine, tangibly embodying a program of instructions executable by the machine to perform method steps for automatically determining an allowable order of changes in a distributed system, said method comprising the steps of:

receiving a request for change;

wherein the request for change describes a task to be done on at least one target system, and a deadline by which a change needs to be completed;

determining existing relationship descriptions between components of the distributed system, wherein the components of the distributed system are implemented on a plurality of managed resources;

transforming acquired relationships into ordered tasks that are linked by temporal ordering constraints which describe when tasks can begin in relation to one another, wherein the temporal ordering constraints are selected from the group consisting of;

Finish-to-Start, Start-to-Start, Finish-to-Finish, and Start-to-Finish;

creating an order of changes taking into account the temporal ordering constraints, wherein creating the order of changes includes determining whether the ordered changes are conflicting and flagging such conflicts and further includes an estimate of the time required to complete a change;

wherein the order of changes transitions the at least one target system from one workable state into another workable state;

building a Task Graph from the order of changes, wherein the Task Graph represents each task to be completed within an overall job;

constructing an Annotated Task Graph by assigning estimated durations to each task within the Task Graph; and

computing a makespan for the overall job represented by the Task Graph, wherein the Annotated Task Graph is returned to an administrator.