Facilitating overall grid environment management by monitoring and distributing grid activity

• US 20060150158A1
• Filed: 01/06/2005
• Published: 07/06/2006
• Est. Priority Date: 01/06/2005
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Hamilton, Rick Allen II, Seaman, James W., Joseph, Joshy, Fellenstein, Craig William
• Granted Patent: US 7,761,557 B2
• Time in Patent Office: Days
• US Class Current: 717/126
• CPC Class Codes: G06F 11/3433 for load management allocation of a server based on load conditions G06F9/505; load rebalancing G06F9/5083; redistributing the load in a network by a load balancer H04L67/1029
• CPC Class Codes: G06F 11/3442 for planning or managing the needed capacity
• CPC Class Codes: G06F 11/3495 for systems
• CPC Class Codes: G06F 2201/815 Virtual middleware or OS functionality using virtual machines to implement generic software techniques for error detection or fault masking G06F11/1484
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

A method, system, and program for facilitating overall grid environment management by monitoring grid activity across disparate grid resources and distributing grid activity to decisional grid modules are provided. A grid workload controller within a computational grid environment monitors real-time grid activity at an application level from multiple disparate grid application environments. The grid workload controller then determines a selection of grid modules within the computational grid environment that require the real-time grid activity to make decisions about the management of the computational grid environment. The grid workload controller distributes the real-time grid activity to the selection of grid modules, wherein the selection of grid modules then make automated decisions within the grid environment to maintain performance requirements.

Claims

1. A computer-implemented method for managing workload within a grid environment, comprising:

monitoring, at a grid workload controller, real-time grid activity at an application level from a plurality of disparate grid application environments within a computational grid environment;

determining a selection of grid modules that require said real-time grid activity to make decisions about the management of said computational grid environment; and

distributing, from said grid workload controller, said real-time grid activity to said selection of grid modules, wherein said selection of grid modules make automated decisions within said grid environment to maintain workload performance requirements within said computational grid environment.

2. The computer-implemented method for managing workload according to claim 1, wherein monitoring, at a grid workload controller, real-time grid activity at an application level from a plurality of disparate grid application environments within a computational grid environment, further comprises:

monitoring, at said grid workload controller, said real-time grid activity received in a plurality of protocols requiring adaptation to a particular protocol used by said grid workload controller.

3. The computer-implemented method for managing workload according to claim 1, wherein monitoring, at a grid workload controller, real-time grid activity at an application level from a plurality of disparate grid application environments within a computational grid environment, further comprises:

receiving, at said grid workload controller, a plurality of types of grid characteristics within said real-time grid activity; and

filtering said plurality of types of grid characteristics within said real-time grid activity according to a plurality of monitoring rules, wherein said plurality of monitoring rules specify a selection of said plurality of types of grid characteristics and a selection of said plurality of disparate grid application environments from which said real-time grid activity is stored.

4. The computer-implemented method for managing workload according to claim 1, wherein monitoring, at a grid workload controller, real-time grid activity at an application level from a plurality of disparate grid application environments within a computational grid environment, further comprises:

interrogating a plurality of manager modules for said real-time grid activity, wherein each of said plurality of manager modules tracks at least one real-time performance characteristic of a grouping of resource nodes allocated to one of said plurality of disparate grid application environments within said computational grid environment.

5. The computer-implemented method for managing workload according to claim 1, wherein determining a selection of grid modules that require said real-time grid activity to make decisions about the management of said computational grid, further comprises:

filtering said real-time grid activity through a plurality of distribution rules set by at least one from among a system administrator and an automated decision controller.

6. The computer-implemented method for managing workload according to claim 1, wherein said real-time grid activity comprise data about at least one from among a receipt of a job request, a pending job offer, a newly scheduled grid job, a currently executing grid job performance, a future job schedule, a current pricing schedule, a future pricing schedule, a current resource availability, and a predicted resource availability.

7. The computer-implemented method for managing workload according to claim 1, wherein said selection of grid modules comprises at least one from among a grid resource allocation module, a grid sell-off module, a grid discounter module, a grid accounting module, and a grid pricing module.

8. The computer-implemented method for managing workload according to claim 1, wherein said grid workload controller is a grid service.

9. A system for managing workload within a grid environment, comprising:

a grid workload controller implemented within a computational grid environment;

said grid workload controller comprising;

means for monitoring real-time grid activity at an application level from a plurality of disparate grid application environments within said computational grid environment;

means for determining a selection of grid modules that require said real-time grid activity to make decisions about the management of said computational grid environment; and

means for distributing said real-time grid activity to said selection of grid modules, wherein said selection of grid modules make automated decisions within said grid environment to maintain workload performance requirements within said computational grid environment.

10. The system for managing workload according to claim 9, wherein said means for monitoring real-time grid activity at an application level from a plurality of disparate grid application environments within said computational grid environment, further comprises:

means for monitoring said real-time grid activity received in a plurality of protocols requiring adaptation to a particular protocol used by said grid workload controller.

11. The system for managing workload according to claim 9, wherein said means for monitoring real-time grid activity at an application level from a plurality of disparate grid application environments within said computational grid environment, further comprises:

means for receiving, at said grid workload controller, a plurality of types of grid characteristics within said real-time grid activity; and

means for filtering said plurality of types of grid characteristics within said real-time grid activity according to a plurality of monitoring rules, wherein said plurality of monitoring rules specify a selection of said plurality of types of grid characteristics and a selection of said plurality of disparate grid application environments from which said real-time grid activity is stored.

12. The system for managing workload according to claim 9, wherein said means for monitoring real-time grid activity at an application level from a plurality of disparate grid application environments within said computational grid environment, further comprises:

means for interrogating a plurality of manager modules for said real-time grid activity, wherein each of said plurality of manager modules tracks at least one real-time performance characteristic of a grouping of resource nodes allocated to one of said plurality of disparate grid application environments within said computational grid environment.

13. The system for managing workload according to claim 9, wherein said means for determining a selection of grid modules that require said real-time grid activity to make decisions about the management of said computational grid, further comprises:

means for filtering said real-time grid activity through a plurality of distribution rules set by at least one from among a system administrator and an automated decision controller.

14. The system for managing workload according to claim 9, wherein said real-time grid activity comprise data about at least one from among a receipt of a job request, a pending job offer, a newly scheduled grid job, a currently executing grid job performance, a future job schedule, a current pricing schedule, a future pricing schedule, a current resource availability, and a predicted resource availability.

15. The system for managing workload according to claim 9, wherein said selection of grid modules comprises at least one from among a grid resource allocation module, a grid sell-off module, a grid discounter module, a grid accounting module, and a grid pricing module.

16. The system for managing workload according to claim 9, wherein said grid workload controller is a grid service.

17. A computer program product, residing on a computer readable medium, for managing workload within a grid environment, comprising:

means for enabling monitoring of real-time grid activity at an application level from a plurality of disparate grid application environments within a computational grid environment;

means for determining a selection of grid modules that require said real-time grid activity to make decisions about the management of said computational grid environment; and

means for controlling distribution of said real-time grid activity to said selection of grid modules.

18. The computer program product for managing workload according to claim 17, wherein said means for enabling monitoring of real-time grid activity at an application level from a plurality of disparate grid application environments within a computational grid environment, further comprises:

means for enabling monitoring of said real-time grid activity received in a plurality of protocols requiring adaptation to a particular protocol used by said grid workload controller.

19. The computer program product for managing workload according to claim 17, wherein said means for enabling monitoring of real-time grid activity at an application level from a plurality of disparate grid application environments within a computational grid environment, further comprises:

means for enabling receipt of a plurality of types of grid characteristics within said real-time grid activity; and

means for filtering said plurality of types of grid characteristics within said real-time grid activity according to a plurality of monitoring rules, wherein said plurality of monitoring rules specify a selection of said plurality of types of grid characteristics and a selection of said plurality of disparate grid application environments from which said real-time grid activity is stored.

20. The computer program product for managing workload according to claim 17, wherein said means for enabling monitoring of real-time grid activity at an application level from a plurality of disparate grid application environments within a computational grid environment, further comprises:

means for controlling interrogation of a plurality of manager modules for said real-time grid activity, wherein each of said plurality of manager modules tracks at least one real-time performance characteristic of a grouping of resource nodes allocated to one of said plurality of disparate grid application environments within said computational grid environment.

21. The computer program product for managing workload according to claim 17, wherein said means for determining a selection of grid modules that require said real-time grid activity to make decisions about the management of said computational grid, further comprises:

means for filtering said real-time grid activity through a plurality of distribution rules set by at least one from among a system administrator and an automated decision controller.

22. The computer program product for managing workload according to claim 17, wherein said means for enabling monitoring of real-time grid activity at an application level from a plurality of disparate grid application environments within a computational grid environment, further comprises

means for enabling monitoring of said real-time grid activity comprising data about at least one from among a receipt of a job request, a pending job offer, a newly scheduled grid job, a currently executing grid job performance, a future job schedule, a current pricing schedule, a future pricing schedule, a current resource availability, and a predicted resource availability.

23. The computer program product for managing workload according to claim 17, wherein said means for determining a selection of grid modules that require said real-time grid activity to make decisions about the management of said computational grid, further comprises:

means for determining said selection of grid modules of at least one from among a grid resource allocation module, a grid group manager module, a grid sell-off module, a grid discounter module, a grid accounting module, and a grid pricing module.

24. The computer program product for managing workload according to claim 17, wherein said computer program product implements a grid service.