Methods and apparatus for implementing a distributed database

• US 9,785,697 B2
• Filed: 06/29/2012
• Issued: 10/10/2017
• Est. Priority Date: 06/29/2012
• Status: Active Grant

• Current Assignee: Telefonaktiebolaget LM Ericsson
• Original Assignee: Telefonaktiebolaget LM Ericsson
• Inventors: Chrcoles Snchez, Jos Maria, Escribano Bullon, Berta Isabel, Ferrando Llopis, Roman
• Primary Examiner(s): PARK, GRACE A
• Application Number:
  US14/410,463
• Publication Number:
  US 20150324444A1
• Time in Patent Office: 1,929 Days
• CPC Class Codes: G06F 16/21 Design, administration or maintenance of databases
• CPC Class Codes: G06F 16/2365 Ensuring data consistency and integrity
• CPC Class Codes: G06F 16/2471 Distributed queries
• CPC Class Codes: G06F 16/278 Data partitioning, e.g. horizontal or vertical partitioning
• CPC Class Codes: G06F 16/285 Clustering or classification

Abstract

According to a first aspect of the present invention there is provided a method of operating a distributed database comprising a plurality of database nodes each storing a plurality of data objects. The method comprises, when executing a database transaction that requires data operations be performed on a plurality of data objects that are distributed between two or more of the database nodes, grouping the data objects that are involved in the database transaction at one of the plurality of database nodes.

Claims

1. A method of operating a distributed database comprising a plurality of database nodes each storing a plurality of data objects, the method comprising:

when executing a database transaction that requires data operations be performed on a plurality of data objects that are distributed between at least two of the database nodes, grouping the data objects that are involved in the database transaction at one of the plurality of database nodes;

the grouping of the data objects at one of the plurality of database nodes comprises storing the data objects that are involved in the database transaction as a data object group at the one of the plurality of database nodes;

associating a group duration threshold timer with the data object group;

starting the group duration threshold timer when the data object group is one of created and modified by one of adding and removing at least one data object to the data object group;

when the group duration threshold timer expires, distributing the data objects that are part of the data object group to at least two of the database nodes; and

discontinuing an association between the data objects that are part of the data object group such that any of the data objects can be one of subsequently moved to a different database and subsequently associated with a different data object group.

2. The method of claim 1, wherein the grouping of the data objects at one of the plurality of database nodes further comprises:

selecting one of the plurality of database nodes that is to store the data objects that are involved in the database transaction; and

for each of the data objects that are involved in the database transaction and that are not stored at the selected database node, copying the data object from a further database node that currently stores the data object to the selected database node.

3. The method of claim 2, further comprising:

for each of the data objects that are involved in the database transaction and that are not stored at the selected database node, when copying the data object from the further database node that currently stores the data object to the selected database node, deleting the data object from the further database node.

4. The method of claim 1, wherein the grouping of the data objects at one of the plurality of database nodes further comprises:

selecting one of the plurality of database nodes that is to store the data objects that are involved in the database transaction; and

for each of the data objects that are involved in the database transaction and that are not stored at the selected database node, copying all of the data objects that are part of the same data object group as the data object from a further database node that currently stores the data object group to the selected database node; and

storing the copied data objects at the selected database node as a data object group.

5. The method of claim 4, further comprising:

for each of the data objects that are involved in the database transaction and that are not stored at the selected database node, when copying all of the data objects that are part of the same data object group as the data object from the further database node that currently stores the data object group to the selected database node, deleting the data object group from the further database node.

6. The method of claim 1, further comprising:

when executing a database transaction that requires creation of at least one data object within the distributed database, grouping the data objects created by the database transaction at one of the plurality of database nodes.

7. The method of claim 6, wherein the grouping of the created data objects at one of the plurality of database nodes comprises:

selecting one of the plurality of database nodes that is to store the created data objects; and

storing the data objects created by the database transaction at the selected database node as a data object group.

8. The method of claim 1, further comprising:

allocating a data object group identifier to the data object group that uniquely identifies the data object group within the distributed database.

9. The method of claim 1, further comprising:

when executing a first database transaction that requires at least one data operation be performed on at least one data object which is part of a data object group, locking the data object group so as to prevent data operations that may be required on any of the data objects that are part of the data object group by a second database transaction.

10. The method of claim 9, wherein the data object group is unlocked so as to allow data operations that may be required by a second database transaction upon one of:

completion of the first database transaction; and

expiry of a lock timer that is started when the data object group is locked and that limits the duration for which the data object group can be locked.

11. The method of claim 1 further comprising:

associating a database transaction interval timer with a data object that is part of a data object group, wherein the database transaction interval timer is reset when the data object is involved in a database transaction; and

when the database transaction interval time expires, allowing the data object to be stored at a different database node to that which stores the data object group.

12. The method of claim 11, wherein, when the database transaction interval time expires, the method further comprises at least one of:

distributing the data object to a database node other than that which stores the data object group; and

discontinuing an association between the data object and the data object group such that the data object can be one of subsequently moved to different database node and subsequently associated with a different data object group.

13. The method of claim 1, wherein the operations of the distributed database are one of executed and ordered by a data distribution manager of the distributed database implemented by one of:

a data distribution node that communicates with the plurality of database nodes;

a client interface node that communicates with database clients and the plurality of database nodes; and

a database node of the plurality of database nodes.

14. An apparatus configured to provide a data distribution management function within a distributed database, the distributed database comprising a plurality of database nodes each storing a plurality of data objects, the apparatus comprising:

a receiver configured to receive requests relating to a database transactions that requires data operations be performed on a plurality of data objects that are distributed between at least two of the database nodes;

a processor configured to;

execute grouping of the data objects that are involved in the database transaction at one of the plurality of database nodes;

generate instructions that cause the database nodes of the distributed database to;

store the data objects that are involved in the database transaction at the one of the plurality of database nodes as a data object group;

associate a group duration threshold timer with the data object group;

start the group duration threshold timer when the data object group is one of created and modified by one of adding and removing at least one data object to the data object group;

upon expiry of the group duration threshold timer, distribute the data objects that are part of the data object group to at least two of the database nodes; and

discontinue an association between the data objects that are part of the data object group such that any of the data objects can be one of subsequently moved to a different database and subsequently associated with a different data object group.

15. The apparatus of claim 14, further comprising a transmitter configured to send the instructions to the database nodes.

16. The apparatus of claim 15, wherein the processor is further configured to generate instructions that cause the database nodes to:

associate a database transaction interval timer with a data object that is part of a data object group, that is reset when the data object is involved in a database transaction, and that, when the database transaction interval time expires, allows the data object to be stored at a different database node to that which stores the data object group.

17. The apparatus of claim 14, wherein the processor is further configured to:

select one of the plurality of database nodes that is to store the data objects that are involved in the database transaction; and

generate instructions that cause the database nodes to copy each of the data objects that are involved in the database transaction and that are not stored at the selected database node from a further database node that currently stores the data object to the selected database node.

18. The apparatus of claim 17,

wherein the processor is further configured to generate instructions that cause the database nodes to delete a data object from the further database node that currently stores the data object when the data object is copied to the selected database node.

19. The apparatus of claim 14, wherein the processor is further configured to:

select one of the plurality of database nodes that is to store the data objects that are involved in the database transaction; and

generate instructions that cause the database nodes to, for each of the data objects that are involved in the database transaction and that are not stored at the selected database node, copy all of the data objects that are part of the same data object group as the data object from a further database node that currently stores the data object group to the selected database node, and to store the copied data objects at the selected database node as a data object group.

20. The apparatus of claim 19, wherein the processor is further configured to generate instructions that cause the database nodes to delete the data object group from the database node that currently stores the data object when copying all of the data objects that are part of the same data object group as the data object from the further database node to the selected database node.

21. The apparatus of claim 14, wherein the receiver is configured to receive requests relating to a database transactions that requires creation of at least one data object within the distributed database, and the processor is configured to generate instructions that cause the database nodes to group the data objects created by the database transaction at one of the plurality of database nodes.

22. The apparatus of claim 21, wherein the processor is further configured to:

select one of the plurality of database nodes that is to store the created data objects; and

generate instructions that cause the selected database node to store the data objects created by the database transaction as a data object group.

23. The apparatus of claim 14, wherein the processor is further configured to allocate a data object group identifier to the data object group that uniquely identifies the data object group within the distributed database.

24. The apparatus of claim 14, wherein, for a first database transaction that requires at least one data operation be performed on at least one data object which is part of a data object group, the processor is further configured to lock the data object group so as to prevent data operations that may be required by a second database transaction on any of the data objects that are part of the data object group.

25. The apparatus of claim 24, wherein the processor is further configured to unlock the data object group so as to allow data operations that may be required by a second database transaction upon one of:

completion of the first database transaction; and

expiry of a lock timer that is started when the data object group is locked and that limits the duration for which the is part of a data object group can be locked.