Apparatus, system, and method for autonomic control of grid system resources

• US 8,332,483 B2
• Filed: 12/15/2003
• Issued: 12/11/2012
• Est. Priority Date: 12/15/2003
• Status: Active Grant

• Current Assignee: JP Morgan Chase Bank, N.A. (JP Morgan Chase & Co)
• Original Assignee: International Business Machines Corporation
• Inventors: Fatula, Joseph John Jr.
• Primary Examiner(s): Donaghue, Larry
• Assistant Examiner(s): TAYLOR, NICHOLAS R
• Application Number:
  US10/735,938
• Publication Number:
  US 20050131993A1
• Time in Patent Office: 3,284 Days
• Field of Search: 709/218, 709/201, 709/205, 709/223, 709/226
• US Class Current: 709/218
• CPC Class Codes: G06F 9/5072 Grid computing

Abstract

An apparatus, system, and method are disclosed for autonomic management of data operations and performance resources on a grid computing system. An autonomic management apparatus includes a monitor module, a policy module, and a regulation module. The monitor module is configured to monitor the grid computing system for a trigger event. The policy module is configured to access one of a plurality of system policies. Each of the plurality of system policies corresponds to an operational control parameter of a system resource of the grid computing system. The regulation module is configured to autonomically regulate the system resource in response to a recognized trigger event according to one of the plurality of system policies. The trigger event may be a prediction trigger event, an initiation trigger event, a regulation trigger event, or a termination trigger event.

Claims

1. An autonomic management apparatus for autonomic management of system resources on a grid computing system, the apparatus comprising:

a storage device storing executable code;

a processor executing the executable code, the executable code comprising;

a monitor module configured to monitor the grid computing system for a predictive trigger event comprising an anticipated change in data flow based on collected historical information;

a policy module configured to access one of a plurality of system policies, each of the plurality of system policies corresponding to an operational control parameter of system resources comprising client processor capacity, client storage capacity, and client memory capacity allocated to the grid computing system, wherein a plurality of clients both provide the system resources and employ the system resources and the plurality of system policies comprises a system prediction policy; and

a regulation module configured to autonomically adjust a system resource allocated to the grid computing system from a client in response to the anticipated change in the data flow and modify a client fee for participation in the grid computing system in response to the adjustment, wherein the client initiates a grid application in response to the anticipated change in data flow, the client fee is proportional to a level of the system resources contributed by the client and a level of grid system demand, and the client fee is based a client use of the grid system selected from the group consisting of instantaneous use, average use, maximum use, minimum use, planned use, reserved use, and peak use.

2. The apparatus of claim 1, wherein the operational control parameter comprises a command to regulate the system resource.

3. The apparatus of claim 1, wherein the regulation module comprises a reservation module configured to reserve the system resource for a grid system operation.

4. The apparatus of claim 1, wherein the regulation module comprises a termination module configured to terminate a reservation of the system resource for a grid system operation.

5. The apparatus of claim 1, wherein the regulation module comprises an arbitration module configured to arbitrate conflicting grid system operations according to an arbitration policy.

6. The apparatus of claim 1, wherein the regulation module comprises a profile module configured to store a system resource profile, the system resource profile identifying the system resource of the client, and the system resource allocated by the client to the grid computing system.

7. The apparatus of claim 1, wherein the plurality of system policies further comprises at least one of a system regulation policy and a system termination policy.

8. A method for autonomic management of system resources on a grid computing system, the method comprising:

monitoring, by use of a processor, the grid computing system for a predictive trigger event comprising an anticipated change in data flow based on collected historical information;

accessing one of a plurality of system policies, each of the plurality of system policies corresponding to an operational control parameter of system resources comprising client processor capacity, client storage capacity, and client memory capacity allocated to the grid computing system, wherein a plurality of clients both provide the system resources and employ the system resources and the plurality of system policies comprises a system prediction policy; and

autonomically adjusting a system resource allocated to the grid computing system from a client in response to the anticipated change in the data flow and modifying a client fee for participation in the grid computing system in response to the adjustment wherein the client initiates a grid application in response to the anticipated change in data flow, the client fee is proportional to a level of the system resources contributed by the client and a level of grid system demand, and the client fee is based a client use of the grid system selected from the group consisting of instantaneous use, average use, maximum use, minimum use, planned use, reserved use, and peak use.

9. The method of claim 8, further comprising reserving the system resource for a grid system operation.

10. The method of claim 8, further comprising terminating a reservation of the system resource for a grid system operation.

18. The method of claim 8, further comprising blocking a potential change in at least one of the system policies according to a threshold corresponding with a subscription criteria.

11. A method for autonomic management of grid system resources on a grid computing system, the method comprising:

monitoring, by use of processor, the grid computing system for a predictive trigger event comprising an anticipated change in data flow based on collected historical information;

accessing one of a plurality of system policies, wherein the plurality of system policies comprises a system prediction policy, each of the plurality of system policies corresponding to an operational control parameter of system resources comprising client processor capacity, client storage capacity, and client memory capacity allocated to the grid computing system, wherein a plurality of clients both provide the system resources and employ the system resources and the plurality of system policies comprises a system prediction policy;

autonomically adjusting a system resource allocated to the grid computing system from a client in response to the anticipated change in the data flow and modifying a client fee for participation in the grid computing system in response to the adjustment, wherein the client initiates a grid application in response to the anticipated change in data flow, the client fee is proportional to a level of the system resources contributed by the client and a level of grid system demand, and the client fee is based a client use of the grid system selected from the group consisting of instantaneous use, average use, maximum use, minimum use, planned use, reserved use, and peak use; and

storing a system resource profile, the system resource profile identifying the system resource of the client.

12. A computer readable storage medium storing executable code executed by a processor to carry out a method for autonomic management of system resources on a grid computing system, the method comprising:

monitoring the grid computing system for a predictive trigger event comprising an anticipated change in data flow based on collected historical information;

accessing one of a plurality of system policies, each of the plurality of system policies corresponding to an operational control parameter of system resources comprising client processor capacity, client storage capacity, and client memory capacity allocated to the grid computing system, wherein a plurality of clients both provide the system resources and employ the system resources and the plurality of system policies comprises a system prediction policy; and

autonomically adjusting a system resource allocated to the grid computing system from a client in response to the anticipated change in the data flow and modifying a client fee for participation in the grid computing system in response to the adjustment, wherein the client initiates a grid application in response to the anticipated change in data flow, the client fee is proportional to a level of the system resources contributed by the client and a level of grid system demand, and the client fee is based a client use of the grid system selected from the group consisting of instantaneous use, average use, maximum use, minimum use, planned use, reserved use, and peak use.

13. The computer readable storage medium of claim 12, wherein the method further comprises reserving the system resource for a grid system operation.

14. The computer readable storage medium of claim 12, wherein the method further comprises terminating a reservation of the system resource for a grid system operation.

15. The computer readable storage medium of claim 12, wherein the method further comprises arbitrating conflicting grid system operations according to an arbitration policy.

16. The computer readable storage medium of claim 12, wherein the method further comprises storing a system resource profile, the system resource profile identifying the system resource of the client, and the system resource allocated by the client to the grid computing system.

17. An apparatus for autonomic management of grid system resources on a grid computing system, the apparatus comprising:

a storage device storing executable code;

a processor executing the executable code, the executable code comprising;

means for monitoring the grid computing system for a predictive trigger event comprising an anticipated change in data flow based on collected historical information;

means for accessing one of a plurality of system policies, each of the plurality of system policies corresponding to an operational control parameter of system resources comprising client processor capacity, client storage capacity, and client memory capacity allocated to the grid computing system, wherein a plurality of clients both provide the system resources and employ the system resources and the plurality of system policies comprises a system prediction policy; and

means for autonomically adjusting a system resource allocated to the grid computing system from a client in response to the anticipated change in the data flow and modifying a client fee for participation in the grid computing system in response to the adjustment, wherein the client initiates a grid application in response to the anticipated change in data flow, the client fee is proportional to a level of the system resources contributed by the client and a level of grid system demand, and the client fee is based a client use of the grid system selected from the group consisting of instantaneous use, average use, maximum use, minimum use, planned use, reserved use, and peak use.