Client and a server and methods thereof for data management relating to shared data storage

• US 9,756,125 B2
• Filed: 06/16/2011
• Issued: 09/05/2017
• Est. Priority Date: 06/16/2011
• Status: Active Grant

• Current Assignee: Telefonaktiebolaget LM Ericsson
• Original Assignee: Telefonaktiebolaget LM Ericsson
• Inventors: Ekeroth, Andreas, Katardjiev, Vladimir, Kristiansson, Johan
• Primary Examiner(s): Nguyen, Dustin
• Assistant Examiner(s): Turriate Gastulo, Juan C
• Application Number:
  US14/125,833
• Publication Number:
  US 20140108516A1
• Time in Patent Office: 2,273 Days
• CPC Class Codes: G06F 11/1658 Data re-synchronization of a redundant component, or initial sync of replacement, additional or spare unit
• CPC Class Codes: G06F 11/2033 switching over of hardware resources
• CPC Class Codes: G06F 11/2038 with a single idle spare processing component
• CPC Class Codes: H04L 67/1095 Replication or mirroring of data, e.g. scheduling or transport for data synchronisation between network nodes

Abstract

A distributed shared data storage such as the distributed shared memory exemplified with the DSM Master can seamlessly migrate (i.e. move) in the runtime between different nodes without affecting the user perceived performance according to embodiments of the present invention. This is achieved by using more than one replica simultaneously operating towards different master databases during the migration. That is, all distributed shared memory replicas can automatically and seamlessly be migrated to another master database, thus making the communication more flexible and efficient.

Claims

1. A method in a client for seamlessly migrating a shared data storage from a first node to a second node, where the shared data storage is accessible to clients accessing the shared data storage during and after the migration, wherein the client comprises a first replica of the shared data storage associated with the first node to be used for operations on the shared data storage, the method comprising:

configuring, by a migration manager in the client, a second replica of the shared data storage associated with the second node;

configuring, by the migration manager in the client, the first and the second replicas to be used for operations simultaneously during the migration from the first node and the second node;

configuring, by the migration manager in the client, a virtual replica encapsulating the first and second replicas internally to provide a single interface towards applications operating on the shared data storage;

receiving information, at the migration manager in the client, that the shared data storage associated with the first node is synchronized with the shared data storage associated with the second node, wherein the synchronization comprises applying operations performed on the first and second replicas to the shared data storage associated with the first node and the shared data storage associated with the second node; and

based on the received information, configuring, by the migration manager in the client, the first and the second replicas to only use the second replica for operations on the shared data storage.

2. The method according to claim 1, wherein each node is identified by an address and the migration manager maintains a lookup table replica mapping an address of the first node with the second node so that each client having a lookup table replica can be notified when a lookup table replica is modified, wherein each lookup table replica is associated with a lookup table master.

3. The method according to claim 1, wherein the received information, at the migration manager, that the shared data storage associated with the first node is synchronized with the shared data storage associated with the second node comprises information that the first and second replicas comprise identical information.

4. The method according to claim 1, wherein merge conflicts on the shared data storage are resolved using Operational Transformations.

5. The method according to claim 1, wherein the migration is triggered based on the load of the first node.

6. The method according to claim 1, wherein at least one of the first and second nodes is a web browser or native application runtime environment.

7. The method according to claim 1, wherein at least one of the first and second nodes is a server.

8. A method in a second node for seamlessly migrating a shared data storage from a first node to the second node, where the shared data storage is accessible to clients during and after the migration, the method comprises:

configuring, by a migration manager in the second node, a first proxy replica of the shared data storage associated with the first node;

configuring, by the migration manager in the second node, a second proxy replica of the shared data storage associated with the second node;

wherein a virtual replica is configured in a client that encapsulates the first and second proxy replicas internally to provide a single interface towards applications that operates on the shared data storage;

configuring, by the migration manager in the second node, a synchronization of the first and the second proxy replicas;

wherein the synchronization is performed by copying data from one of the first and second proxy replicas to other of the first and second proxy replicas upon receiving an update by the one of the first and second proxy replicas regarding an operation performed by the client or the other of the first and second proxy replicas; and

updating a lookup table replica maintained, by the migration manager in the second node, with information that the shared data storage associated with the second node can be used in addition to the shared data storage associated with the first node, wherein the lookup table replica is associated with a lookup table master which in turn is associated with lookup table replicas of other nodes and clients.

9. The method according to claim 8, wherein each node is identified by an address and the migration manager is controlling a lookup table replica mapping an address of the first node with the second node so that each client can be notified via a respective lookup table replica when the lookup table replica of the migration manager of the second node is updated.

10. The method according to claim 8, wherein merge conflicts on the shared data storage are resolved using Operational Transformations.

11. The method according to claim 10, wherein the method further comprises preventing the first and second proxy replicas from diverging by more than one state, by acquiring and releasing a lock via a shared mutex for each incoming operation, where only a proxy replica client that has acquired the shared mutex can process incoming operations.

12. The method according to claim 8, wherein at least one of the first and second nodes is a web browser or native application runtime environment.

13. The method according to claim 8, wherein at least one of the first and second nodes is a server.

14. A client of an end user device which includes a processor and a memory for seamlessly migrating a shared data storage from a first node to a second node, where the shared data storage is accessible to clients accessing the shared data storage during and after the migration, wherein the client comprises:

a first replica of the shared data storage associated with the first node to be used for operations on the shared data storage; and

a migration manager that;

configures a second replica of the shared data storage associated with the second node;

configures the first and the second replicas to be used for operations simultaneously during the migration from the first node and the second node;

configures a virtual replica encapsulating the first and second replicas internally to provide a single interface towards applications operating on the shared data storage; and

based on receiving information that the shared data storage associated with the first node is synchronized with the shared data storage associated with the second node, wherein the synchronization comprises applying operations performed on the first and second replicas to the shared data storage associated with the first node and the shared data storage associated with the second node, configures the first and the second replicas to only use the second replica for operations on the shared data storage.

15. The client according to claim 14, wherein the migration manager maintains a lookup table replica mapping an address of the first node with the second node, so that each client having a lookup table replica can be notified when a lookup table replica is modified, wherein each lookup table replica is configured to be associated with a lookup table master.

16. The client according to claim 14, wherein the received information that the shared data storage associated with the first node should be synchronized with the shared data storage associated with the second node comprises information that the first and second replicas comprise identical information.

17. The client according to claim 14, wherein the client is configured to resolve merge conflicts on the shared data storage using Operational Transformations.

18. The client according to claim 14, wherein the migration manager triggers the migration based on the load of the first node.

19. The client according to claim 14, wherein at least one of the first and second nodes is a web browser or native application runtime environment.

20. The client according to claim 14, wherein at least one of the first and second nodes is a server.

21. A second node for seamlessly migrating a shared data storage from a first node to the second node wherein the first node and the second node each include a processor and a memory, where the shared data storage is accessible to clients during and after the migration, wherein the second node comprises a migration manager that configures:

a first proxy replica of the shared data storage associated with the first node,

a second proxy replica of the shared data storage associated with the second node,

wherein a virtual replica is configured in a client that encapsulates the first and second proxy replicas internally to provide a single interface towards applications that operates on the shared data storage,

a synchronization of the first and the second proxy replicas,

wherein the synchronization is performed by copying data from one of the first and second proxy replicas to other of the first and second proxy replicas upon receiving an update by the one of the first and second proxy replicas regarding an operation performed by the client or the other of the first and second proxy replicas, and

updates a lookup table replica maintained by the migration manager with information that the shared data storage associated with the second node can be used in addition to the shared data storage associated with the first node, wherein the lookup table replica is configured to be associated with a lookup table master which in turn is configured to be associated with lookup table replicas of other nodes and clients.

22. The second node according to claim 21, wherein the migration manager controls a lookup table replica mapping the addresses of the first node with the second node so that each client can be notified via a respective lookup table replica when the lookup table replica of the migration manager of the second node is updated.

23. The second node according to claim 21, wherein the second node is configured to resolve merge conflicts on the shared data storage using Operational Transformations.

24. The second node according to claim 23, wherein the second node is configured to prevent the first and second proxy replicas to diverge more than one state by acquiring and releasing a lock via a shared mutex for each incoming operation, where only the proxy replica client that has acquired the shared mutex can process incoming operations.

25. The second node according to claim 21, wherein at least one of the first and second nodes is a web browser or native application runtime environment.

26. The second node according to claim 21, wherein at least one of the first and second nodes is a server.