Grid quorum

• US 7,299,229 B2
• Filed: 04/25/2003
• Issued: 11/20/2007
• Est. Priority Date: 04/25/2003
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Doyle, Ronald P., Kaminsky, David Louis
• Primary Examiner(s): Wong; Don
• Assistant Examiner(s): Black; Linh
• Application Number:
  US10/424,034
• Publication Number:
  US 20040215614A1
• Time in Patent Office: 1,670 Days
• Field of Search: 707 1-1041
• US Class Current: 1/1
• CPC Class Codes: G06F 9/50 Allocation of resources, e.g. of the central processing unit [CPU]
• CPC Class Codes: G06F 9/5072 Grid computing
• CPC Class Codes: G06F 9/526 Mutual exclusion algorithms
• CPC Class Codes: Y10S 707/99931 Database or file accessing
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99938 Concurrency, e.g. lock management in shared database

Abstract

The present invention is a grid quorum system, method and apparatus. In a cluster of resources in a computing grid, a resource locking method can include acquiring a temporally limited lock on a grid service in the computing grid. Upon expiration of the temporally limited lock, a renewal of the temporally limited lock can be requested. Subsequently, the temporally limited lock can be renewed if a renewal has been granted by the grid service in response to the request. Notably, the renewing step can include determining whether the cluster has been partitioned into a plurality of sub-clusters. If the cluster has been partitioned, a parent sub-cluster can be identified and the temporally limited lock can be renewed only if a quorum exists in the parent sub-cluster.

Claims

1. In a cluster of resources in a computing grid, a resource locking method comprising:

acquiring a temporally limited lock on a grid service in the computing grid;

upon expiration of said temporally limited lock, requesting a renewal of said temporally limited lock; and

,

renewing said temporally limited lock if a renewal has been granted by said grid service in response to said request, wherein said acquiring step comprises the steps of;

determining whether a resource within the cluster already holds a temporally limited lock to said grid service; and

,

if no resource can be determined to hold a temporally limited lock to said grid service, competing with other resources in the cluster to acquire said temporally limited lock to said grid service.

2. The method of claim 1, wherein said renewing step comprises the steps of:

determining whether the cluster has been partitioned into a plurality of sub-clusters; and

,

if the cluster has been partitioned, identifying a parent sub-cluster and renewing said temporally limited lock only if a quorum exists in said parent sub-cluster.

3. The method of claim 2, further comprising the step of computing said quorum by determining whether a number of resources in said parent sub-cluster exceeds more than half of all resources in the cluster.

4. The method of claim 2, further comprising the step of computing said quorum by determining whether a number of resources in said parent sub-cluster comprises at least half of all resources in the cluster.

5. The method of claim 2, further comprising the step of releasing said temporally limited lock if a quorum does not exist in said parent sub-cluster.

6. The method of claim 2, further comprising the step of allowing said temporally limited lock to expire if a quorum does not exist in said parent sub-cluster.

7. The method of claim 1, wherein said step of competing comprises voting to obtain said temporally limited lock.

8. The method of claim 1, wherein said step of competing comprises racing to obtain said temporally limited lock.

9. A machine readable storage having stored thereon a computer program for resource locking in a cluster of resources in a computing grid, said computer program comprising a routine set of instructions for causing the machine to perform the steps of:

acquiring a temporally limited lock on a grid service in the computing grid;

upon expiration of said temporally limited lock, requesting a renewal of said temporally limited lock; and

,

renewing said temporally limited lock if a renewal has been granted by said grid service in response to said request, wherein said acquiring step comprises the steps of;

determining whether a resource within the cluster already holds a temporally limited lock to said grid service; and

,

if no resource can be determined to hold a temporally limited lock to said grid service, competing with other resources in the cluster to acquire said temporally limited lock to said grid service.

10. The machine readable storage of claim 9, wherein said renewing step comprises the steps of:

determining whether the cluster has been partitioned into a plurality of sub-clusters; and

,

if the cluster has been partitioned, identifying a parent sub-cluster and renewing said temporally limited lock only if a quorum exists in said parent sub-cluster.

11. The machine readable storage of claim 10, further comprising the step of computing said quorum by determining whether a number of resources in said parent sub-cluster exceeds more than half of all resources in the cluster.

12. The machine readable storage of claim 10, further comprising the step of computing said quorum by determining whether a number of resources in said parent sub-cluster comprises at least half of all resources in the cluster.

13. The machine readable storage of claim 10, further comprising the step of releasing said temporally limited lock if a quorum does not exist in said parent sub-cluster.

14. The machine readable storage of claim 10, further comprising the step of allowing said temporally limited lock to expire if a quorum does not exist in said parent sub-cluster.

15. The machine readable storage of claim 9, wherein said step of competing comprises voting to obtain said temporally limited lock.

16. The machine readable storage of claim 9, wherein said step of competing comprises racing to obtain said temporally limited lock.