SYSTEM AND METHOD FOR SUPPORTING SERVICE LEVEL QUORUM IN A DATA GRID CLUSTER

• US 20160119197A1
• Filed: 12/30/2015
• Published: 04/28/2016
• Est. Priority Date: 01/28/2011
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Lee, Robert H., Purdy, Cameron, Falco, Mark, Glezyer, Gene
• Granted Patent: US 10,122,595 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 11/1425 by reconfiguration of node membership
• CPC Class Codes: G06F 2209/505 Clust
• CPC Class Codes: G06F 9/5061 Partitioning or combining of resources
• CPC Class Codes: H04L 41/5054 Automatic deployment of services triggered by the service manager, e.g. service implementation by automatic configuration of network components
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46
• CPC Class Codes: H04L 67/51 Discovery or management thereof, e.g. service location protocol [SLP] or web services

Abstract

A system and method is described for use with a data grid cluster, for supporting service level quorum in the data grid cluster. The data grid cluster includes a plurality of cluster nodes that support performing at least one service action. A quorum policy, defined in a cache configuration file associated with the data grid cluster, can specify a minimum number of service members that are required in the data grid cluster for performing the service action. The data grid cluster uses the quorum policy to determine whether the service action is allowed to be performed, based on a present state of the plurality of cluster nodes in the data grid cluster.

Claims

1-13. -13. (canceled)

14. A system for supporting service level quorum in a data grid cluster, comprising:

a plurality of cluster nodes in the data grid cluster, wherein the data grid cluster supports performing a plurality of services including a root cluster service performed by all of the plurality of cluster nodes, and a first service performed by a first subset of said plurality of cluster nodes;

a root cluster service quorum policy defined in a cache configuration file associated with the data grid cluster, wherein the root cluster service quorum policy specifies a minimum number of qualified root cluster service members that are required for performing root cluster service actions of said root cluster service;

a first service quorum policy defined in the cache configuration file associated with the data grid cluster, wherein the first service quorum policy specifies a minimum number of qualified first service members required for performing first service actions of said first service;

wherein the data grid cluster uses the root cluster service quorum policy to determine whether root service actions are allowed to be performed based on how many qualified root cluster service members are operational in said plurality of cluster nodes; and

wherein the data grid cluster uses the first service quorum policy to determine whether first service actions are allowed to be performed based on how many cluster nodes are operational in said first subset of said plurality of cluster nodes.

15. The system according to claim 14, further comprising:

a pluggable interface that allows users to implement at least one custom action policy defining a custom action controlling behaviors of the data grid cluster, wherein the data grid cluster operates to use a factory class to create the at least one custom action policy.

16. The system according to claim 15, wherein:

the first service quorum policy contains a reference to the at least one custom action policy.

17. The system according to claim 14, wherein:

the root cluster service quorum policy defines a timeout survivor quorum that specifies a minimum number of qualified root cluster service members that must remain in the data grid cluster when the root cluster service is terminating suspect members, wherein a cluster node becomes a suspected cluster member when it has not responded to network communications.

18. The system according to claim 14, wherein the first service is a partitioned cache service and first service quorum policy defines at least one of:

a distribution quorum that specifies a minimum number of storage-enabled members of the partitioned cache service that must be present before the partitioned cache service is allowed to perform partition distribution;

a restore quorum that specifies a minimum number of storage-enabled members of the partitioned cache service that must be present before the partitioned cache service is allowed to restore lost primary partitions from backup;

a read quorum that specifies a minimum number of storage-enabled members of the partitioned cache service that must be present in order to process read requests, wherein read requests do not mutate the state or contents of a cache; and

a write quorum that specifies a minimum number of storage-enabled members of the partitioned cache service that must be present in order to process write requests, wherein write requests may mutate the state or contents of the cache.

19. The system of claim 14, wherein the first service is a proxy service and the first service quorum policy:

defines a connection quorum that specifies a minimum number of proxy service members that must be available before the proxy service can allow client connections.

20. The system of claim 14, wherein:

the first service quorum policy is defined in a scheme element in the cache configuration file, when the scheme element supports scheme reference.

21. The system of claim 14, further comprising:

at least one custom action policy defining a custom action controlling behaviors of the data grid cluster, wherein the at least one custom policy is defined based on at least one of

a particular time,

availability of a particular resource, and

an external event.

22. The system of claim 21, wherein:

the first service quorum policy refers to the at least one custom action policy.

23. A method for supporting service level quorum in a data grid cluster, comprising a plurality of cluster nodes wherein the data grid cluster supports performing a plurality of services including a root cluster service performed by all of the plurality of cluster nodes and a first service performed by a first subset of said plurality of cluster nodes, the method comprising:

providing a root cluster service quorum policy in a cache configuration file associated with data grid cluster, wherein the root cluster service quorum policy specifies a minimum number of qualified root cluster service members that are required for performing root cluster service actions of said root cluster service;

providing a first service quorum policy defined in the cache configuration file associated with the data grid cluster, wherein the first service quorum policy specifies a minimum number of qualified first service members that are required for performing first service actions of said first service;

using the root cluster service quorum policy to determine whether root cluster service actions are allowed to be performed based on how many qualified root cluster service members are operational in said plurality of cluster nodes; and

using the first service quorum policy to determine whether first service actions are allowed to be performed based on how many cluster nodes are operational in said first subset of said plurality of cluster nodes.

24. The method of claim 23, further comprising:

providing a pluggable interface that allows users to implement a custom action policy in order to control behaviors of the data grid cluster; and

using a factory class to create the custom action policy.

25. The method of claim 24, further comprising:

making reference from the first service quorum policy to the custom action policy.

26. The method of claim 23, further comprising:

defining, via the root cluster service quorum policy, a timeout survivor quorum that specifies a minimum number of cluster nodes that must remain in the data grid cluster when a root cluster service is terminating suspect members, wherein a cluster node becomes a suspected cluster member when it has not responded to network communications.

27. The method of claim 23, wherein the first service is a partitioned cache service, and wherein the method further comprises defining, via the first service quorum policy, at least one of:

a distribution quorum that specifies a minimum number of storage-enabled members of the partitioned cache service that must be present before the partitioned cache service is allowed to perform partition distribution;

a restore quorum that specifies a minimum number of storage-enabled members of the partitioned cache service that must be present before the partitioned cache service is allowed to restore lost primary partitions from backup;

a read quorum that specifies a minimum number of storage-enabled members of the partitioned cache service that must be present in order to process read requests, wherein read requests do not mutate the state or contents of a cache; and

a write quorum that specifies a minimum number of storage-enabled members of the partitioned cache service that must be present in order to process write requests, wherein write requests may mutate the state or contents of the cache.

28. The method of claim 23, wherein the first service is a proxy service further comprising:

defining, via the first service quorum policy, a connection quorum that specifies a minimum number of proxy service members that must be available before the proxy service can allow client connections.

29. The method of claim 23, further comprising:

defining the first service quorum policy in a scheme element in the cache configuration file, when the scheme element supports scheme reference.

30. The method of claim 23, further comprising:

defining a custom action policy in order to control behaviors of the data grid cluster, wherein the custom action policy is based on at least one of, a particular time, availability of a particular resource, and an external event.

31. The method of claim 23, further comprising:

making reference from the first service quorum policy to the custom action policy.

32. A non-transitory machine readable medium having instructions stored thereon for supporting service level quorum in a data grid cluster comprising a plurality of cluster nodes wherein the data grid cluster supports performing a plurality of services including a root cluster service performed by all of the plurality of cluster nodes and a first service performed by a first subset of said plurality of cluster nodes, which instructions, when executed cause the data grid cluster to perform steps comprising:

providing a root cluster service quorum policy in a cache configuration file associated with data grid cluster, wherein the root cluster service quorum policy specifies a minimum number of qualified root cluster service members that are required for performing root cluster service actions of said root cluster service;

providing a first service quorum policy defined in the cache configuration file associated with the data grid cluster, wherein the first service quorum policy specifies a minimum number of qualified first service members that are required for performing first service actions of said first service;

using the root cluster service quorum policy to determine whether root cluster service actions are allowed to be performed based on how many qualified root cluster service members are operational in said plurality of cluster nodes; and

using the first service quorum policy to determine whether first service actions are allowed to be performed based on how many cluster nodes are operational in said first subset of said plurality of cluster nodes.

33. The non-transitory machine readable medium of claim 32 having further instructions stored thereon for supporting service level quorum in a data grid cluster wherein the first service is a partitioned cache service, which further instructions, when executed cause the data grid cluster to perform further steps comprising defining, via the first service quorum policy, at least one of:

a distribution quorum that specifies a minimum number of storage-enabled members of the partitioned cache service that must be present before the partitioned cache service is allowed to perform partition distribution;

a restore quorum that specifies a minimum number of storage-enabled members of the partitioned cache service that must be present before the partitioned cache service is allowed to restore lost primary partitions from backup;

a read quorum that specifies a minimum number of storage-enabled members of the partitioned cache service that must be present in order to process read requests, wherein read requests do not mutate the state or contents of a cache; and

a write quorum that specifies a minimum number of storage-enabled members of the partitioned cache service that must be present in order to process write requests, wherein write requests may mutate the state or contents of the cache.