Automatic resource group formation and maintenance in a high availability cluster configuration

• US 6,038,677 A
• Filed: 03/31/1997
• Issued: 03/14/2000
• Est. Priority Date: 03/31/1997
• Status: Expired due to Term

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Arendt, James Wendell, Strong, Hovey Raymond, Lawlor, Francis D.
• Primary Examiner(s): WRIGHT, NORMAN M
• Application Number:
  US08/829,569
• Time in Patent Office: 1,079 Days
• Field of Search: 395/182.02, 395/182.09, 395/182.1, 395/182.11, 395/182.13, 395/182.14, 395/182.16, 395/182.18, 707/202, 707/203, 714/4, 714/9, 714/12, 714/2, 714/3, 709/220, 709/226
• US Class Current: 714/4.1
• CPC Class Codes: G06F 11/142 Reconfiguring to eliminate the error group management mechanisms in a peer-to-peer network H04L67/1044
• CPC Class Codes: G06F 11/20 using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
• CPC Class Codes: G06F 9/5016 the resource being the memory

Abstract

A method and apparatus that automatically generates and maintains resource groups for a clustered computer network configuration. Resource groups are said to be generated "automatically" because the system administrator preferably is not directly involved with defining what resources go within a particular group. Rather, the administrator merely identifies a set of resources that must be collocated with a given application in the event of a failure of a computer on which the application is then executing. One or more resource groups are then automatically generated using a set of collocation "constraints" or rules. A first collocation constraint preferably enforces any user-defined collocations for a given application, and a second constraint collocates disk partition resources residing on the same physical disk. A resource group generated is this manner ensures effective fault-tolerant operation.

Claims

1. A method of using cluster-wide resources in a computer network having a plurality of nodes to provide uninterrupted processing of software applications, comprising the steps of:

for each of a plurality of applications whose processing is to be uninterrupted, having a user identify a set of resources that must be collocated with each such application in the event of a failure of a computer on which the application is then executing;

automatically generating one or more resource groups, each of which includes one or more of the plurality of applications and their related resources; and

upon a given occurrence, transferring control of the resources of a resource group to a backup computer.

2. The method as described in claim 1 wherein the set of resources includes resources selected from the group consisting of applications, IP addresses and disk partitions.

3. The method as described in claim 1 wherein the step of automatically generating one or more resource groups computationally assigns each resource to a resource group in a predetermined manner such that the resource groups are as small as possible and each set of resources required to be collocated is completely contained within a given resource group.

4. The method as described in claim 1 wherein the step of automatically generating one or more resource groups includes enforcing a set of collocation constraints.

5. The method as described in claim 4 wherein the collocation constraints include user-defined collocations for a given application.

6. The method as described in claim 4 wherein the collocation constraints collocate disk partition resources residing on a physical disk.

7. A method of configuring cluster-wide resources in a computer network having a plurality of nodes to facilitate uninterrupted processing of software applications, comprising the steps of:

for each of a plurality of applications whose processing is to be uninterrupted, having a user identify a set of resources that must be collocated with each such application in the event of a failure of a computer on which the application is then executing; and

automatically generating one or more resource groups, each of the one or more resource groups comprising one or more of the plurality of applications and their related resources.

8. The method as described in claim 7 wherein the set of resources includes resources selected from the group consisting of applications, IP addresses and disk partitions.

9. The method as described in claim 7 wherein the step of automatically generating one or more resource groups includes enforcing a set of collocation constraints.

10. The method as described in claim 9 wherein the collocation constraints include user-defined collocations for a given application.

11. The method as described in claim 9 wherein the collocation constraints collocate disk partition resources residing on a physical disk.

12. A method of managing cluster-wide resources in a computer network having a plurality of nodes to facilitate uninterrupted processing of software applications, comprising the steps of:

for each of a plurality of applications whose processing is to be uninterrupted, having a user identify a set of resources that must be collocated with each such application in the event of a failure of a computer on which the application is then executing;

automatically generating one or more resource groups, each of which includes one or more of the plurality of applications and their related resources;

in response to any reconfiguration of a cluster-wide resource, evaluating whether the reconfiguration impacts any resource group; and

if the reconfiguration impacts any resource group, notifying the user.

13. The method as described in claim 12 wherein the notifying step provides the user with information indicating how the resource group is impacted.

14. The method as described in claim 12 wherein the set of resources includes resources selected from the group consisting of applications, IP addresses and disk partitions.

15. A computer program product in a computer-readable medium for use in a data processing system to manage cluster-wide resources in a computer network, comprising:

first program data means responsive to user input for identifying a set of resources that must be collocated with an application in the event of a failure of a computer on which the application is then executing; and

second program data means responsive to the user input and a set of collocation constraints for generating one or more resource groups.

16. The computer program product as in claim 15 wherein the second program data means implements user-defined collocations for a given application.

17. The computer program product as in claim 15 wherein the second program data means collocates disk partition resources residing on a physical disk.

18. In a computer network having a primary computer and a backup computer, a method of using cluster-wide resources to provide uninterrupted processing of software applications, comprising the steps of:

for each of a plurality of applications whose processing is to be uninterrupted, having a user identify a set of resources that must be collocated with each such application in the event of a failure of the primary computer;

automatically generating one or more resource groups, each of which includes one or more of the plurality of applications and their related resources; and

upon a given occurrence at the primary computer, transferring control of the resources of a resource group to the backup computer.