SCALE ACROSS IN A GRID COMPUTING ENVIRONMENT

• US 20080256223A1
• Filed: 04/13/2007
• Published: 10/16/2008
• Est. Priority Date: 04/13/2007
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Govindaraju, Rama K., Skovira, Joseph F., Chan, Waiman
• Granted Patent: US 7,987,467 B2
• Time in Patent Office: Days
• US Class Current: 709/223
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/1008 based on parameters of servers, e.g. available memory or workload monitoring of computer activity G06F11/30
• CPC Class Codes: H04L 67/1029 using data related to the state of servers by a load balancer

Abstract

A method, grid computing environment, and computer readable medium for managing available resources in a grid computing environment are disclosed. The method includes determining, when a new job is submitted, if a first of the computing clusters has sufficient available resources to run the new job. If the first computing cluster does not have sufficient available resources to run the new job, the method further includes determining if a second of the computing clusters has sufficient available resources to run a first job that is currently running on the first computing cluster. If the second computing cluster has sufficient available resources to run the first job, the first job is migrated to the second computing cluster.

Claims

1. A method for managing available resources in a grid computing environment that includes a plurality of computing clusters, the method comprising the steps of:

when a new job is submitted, determining if a first of the computing clusters has sufficient available resources to run the new job;

if the first computing cluster does not have sufficient available resources to run the new job, determining if a second of the computing clusters has sufficient available resources to run a first job that is currently running on the first computing cluster; and

if the second computing cluster has sufficient available resources to run the first job, migrating the first job to the second computing cluster.

2. The method of claim 1, wherein the migrating step comprises the sub-steps of:

canceling the first job at the first computing cluster so as to free resources of the first computing cluster; and

running the first job on the second computing cluster.

3. The method of claim 2, wherein the migrating step further comprises the sub-step of running the new job on the first computing cluster using at least some of the resources of the first computing cluster that were freed by canceling the first job.

4. The method of claim 2,

wherein the migrating step further comprises the sub-step of setting a checkpoint for the first job before canceling the first job at the first computing cluster, and

the sub-step of running the first job on the second computing cluster comprises using the checkpoint to restart the first job on the second computing cluster from the point at which it was canceled.

5. The method of claim 1, further comprising the step of monitoring available resources of each of the computing clusters of the grid computing environment.

6. The method of claim 1, further comprising the step of:

after migrating the first job to the second computing cluster, if the first computing cluster still does not have sufficient available resources to run the new job, migrating a second job that is currently running on the first computing cluster to the second computing cluster or to a third of the computing clusters.

7. The method of claim 1, wherein the first job is chosen for migration based on job priorities.

8. The method of claim 1, further comprising:

determining, in response to at least one of a job completing, a job being cancelled, and a detected resource change, a set of potential job migrations;

determining that the second of the computing clusters does not have sufficient available resources to run the first job;

selecting, in response to determining that the second of the computing clusters does not have sufficient available resources to run the first job, a subset of job migrations from the set of potential job migrations; and

applying the subset of job migrations to the second of the computing clusters to free up enough resources on the second of the computing clusters to run the first job.

9. A computer-readable medium encoded with a program for managing available resources in a grid computing environment that includes a plurality of computing clusters, said program containing instructions for performing the steps of:

when a new job is submitted, determining if a first of the computing clusters has sufficient available resources to run the new job;

if the first computing cluster does not have sufficient available resources to run the new job, determining if a second of the computing clusters has sufficient available resources to run a first job that is currently running on the first computing cluster; and

if the second computing cluster has sufficient available resources to run the first job, migrating the first job to the second computing cluster.

10. The computer-readable medium of claim 9, wherein the migrating step comprises the sub-steps of:

canceling the first job at the first computing cluster so as to free resources of the first computing cluster; and

running the first job on the second computing cluster.

11. The computer-readable medium of claim 10, wherein the migrating step further comprises the sub-step of running the new job on the first computing cluster using at least some of the resources of the first computing cluster that were freed by canceling the first job.

12. The computer-readable medium of claim 10,

wherein the migrating step further comprises the sub-step of setting a checkpoint for the first job before canceling the first job at the first computing cluster, and

the sub-step of running the first job on the second computing cluster comprises using the checkpoint to restart the first job on the second computing cluster from the point at which it was canceled.

13. The computer-readable medium of claim 9, wherein said program further contains instructions for performing the step of monitoring available resources of each of the computing clusters of the grid computing environment.

14. The computer-readable medium of claim 9, wherein said program further contains instructions for performing the step of:

after migrating the first job to the second computing cluster, if the first computing cluster still does not have sufficient available resources to run the new job, migrating a second job that is currently running on the first computing cluster to the second computing cluster or to a third of the computing clusters.

15. The computer-readable medium of claim 9, wherein said program further contains instructions for performing the step of:

determining, in response to at least one of a job completing, a job being cancelled, and a detected resource change, a set of potential job migrations;

determining that the second of the computing clusters does not have sufficient available resources to run the first job;

selecting, in response to determining that the second of the computing clusters does not have sufficient available resources to run the first job, a subset of job migrations from the set of potential job migrations; and

applying the subset of job migrations to the second of the computing clusters to free up enough resources on the second of the computing clusters to run the first job.

16. A grid computing environment including a plurality of computing clusters, the grid computing environment comprising:

a scale across module for, when a new job is submitted, determining if a first of the computing clusters has sufficient available resources to run the new job,

wherein if the scale across module determines that the first computing cluster does not have sufficient available resources to run the new job, the scale across module determines if a second of the computing clusters has sufficient available resources to run a first job that is currently running on the first computing cluster, and if so migrates the first job to the second computing cluster.

17. The grid computing environment of claim 16, wherein the scale across module comprises a job migrator for canceling the first job at the first computing cluster so as to free resources of the first computing cluster, and moving the first job to the second computing cluster.

18. The grid computing environment of claim 17, wherein after the scale across module moves the first job to the second computing cluster, the scale across module runs the new job on the first computing cluster using at least some of the resources of the first computing cluster that were freed by canceling the first job.

19. The grid computing environment of claim 17, wherein the scale across module further comprises:

a checkpoint module for setting a checkpoint for the first job before canceling the first job at the first computing cluster; and

a job restart module for using the checkpoint to restart the first job on the second computing cluster from the point at which it was canceled.

20. The grid computing environment of claim 16, wherein, after migrating the first job to the second computing cluster, if the first computing cluster still does not have sufficient available resources to run the new job, the scale across module migrates a second job that is currently running on the first computing cluster to the second computing cluster or to a third of the computing clusters.