MOVING DATA BETWEEN PARTITIONS

• US 20150347401A1
• Filed: 05/29/2014
• Published: 12/03/2015
• Est. Priority Date: 05/29/2014
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Raghavan, Ananth, Eadon, George, Kumar, Ramesh, Baer, Hermann, Sridharan, Suresh
• Granted Patent: US 9,965,497 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/214 Database migration support

Abstract

Techniques are provided for moving data between partitions. Such a process may be performed without requiring any locks that block transactions that target a partition from being executed. Instead, such transactions may proceed while a move operation is being performed. The move operation involves copying data from the targeted partition to another partition that is hidden from (or “invisible” to) those transactions that attempt to read from or write to the partition. During the move operation, changes that are made to the partition are also reflected in a journal. Eventually, the changes reflected in the journal are drained and applied to the hidden partition. Once the partition and the hidden partition are synchronized, the identities of the partitions are swapped so that future transactions will target the previously-hidden partition instead of the previously-viewable partition.

Claims

1. A method comprising:

receiving an instruction to move a partition to a new storage location, to split the partition into two or more partitions, or to merge a plurality of partitions that includes the partition into a single partition;

in response to receiving the instruction and while allowing a plurality of transactions to modify data reflected in the partition, performing an operation that moves, splits, or merges the partition;

wherein the method is performed by one or more computing devices.

2. The method of claim 1, wherein performing the operation comprises:

creating a new partition that a plurality of transactions cannot access;

copying data, from the partition that the plurality of transactions can access, into the new partition.

3. The method of claim 2, wherein performing the operation further comprises:

generating journaling metadata that indicates that transactions are to store change information, caused by the transactions, in a journal;

after generating the journaling metadata, generating cooperation metadata that indicates that subsequent transactions are to check the journal and apply changes, caused by the subsequent transactions, to the new partition.

4. The method of claim 3, further comprising:

in response to determining a particular time after the journaling metadata is generated and before the cooperation metadata is generated, waiting for one or more transactions that were pending at the particular time to commit before copying the data.

5. The method of claim 3, further comprising:

after generating the cooperation metadata, identifying a particular change in the journal and applying the particular change to the new partition.

6. The method of claim 5, wherein:

identifying and applying are performed by a process that is executing a data definition language (DDL) transaction that began prior to creating the new partition;

the process performs copying the data, generating the journaling data, and generating the cooperation metadata.

7. The method of claim 3, further comprising:

in response to determining a particular time that the journal does not contain any change entries that have not been applied to the new partition, waiting for one or more transactions that were pending at the particular time to commit before exchanging identities of the partition and the new partition.

8. The method of claim 3, further comprising:

after generating the cooperation metadata, exchanging identities of the partition and the new partition.

9. The method of claim 3, further comprising invalidating one or more first cursors after creating the journal and invalidating one or more second cursors after generating the cooperation metadata.

10. The method of claim 2, wherein performing the operation comprises:

determining whether a local index exists for the partition;

in response to determining that a local index exists for the partition, generating a new local index for the new partition.

11. The method of claim 2, further comprising:

generating a plurality of mappings, each of which maps a location identifier for the partition with a location identifier for the new partition.

12. The method of claim 2, further comprising:

determining that a global index exists for the partition;

in response to receiving a query;

identifying an entry in the global index,

identifying a first location identifier that is indicated in the entry,

based on a mapping in a plurality of mappings, identifying a second location identifier that corresponds to the first location identifier,

using the second location identifier to retrieve data from the new partition.

13. A method comprising:

storing a plurality of mappings, each of which maps a location identifier for a first database object with a location identifier for a second database object that is different than the first database object;

in response to receiving a query;

identifying an entry in an index,

identifying a first location identifier that is indicated in the entry,

based on a mapping in the plurality of mappings, identifying a second location identifier that corresponds to the first location identifier,

using the second location identifier to retrieve data from the second database object;

wherein the method is performed by one or more computing devices.

14. The method of claim 13, wherein the first database object is a first partition and the second database object is a second partition.

15. The method of claim 13, further comprising:

identifying a second entry in the index;

identifying a particular location identifier that is indicated in the second entry,

determining whether the particular location identifier is for the first database object or the second database object;

in response to determining that the particular location identifier is for the second database object, then using the particular location identifier to retrieved second data from the second database object.

16. One or more non-transitory storage media storing instructions which, when executed by one or more computing devices, cause:

receiving an instruction to move a partition to a new storage location, to split the partition into two or more partitions, or to merge a plurality of partitions that includes the partition into a single partition;

in response to receiving the instruction and while allowing a plurality of transactions to modify data reflected in the partition, performing an operation that moves, splits, or merges the partition;

wherein the method is performed by one or more computing devices.

17. The one or more non-transitory storage media of claim 16, wherein performing the operation comprises:

creating a new partition that a plurality of transactions cannot access;

copying data, from the partition that the plurality of transactions can access, into the new partition.

18. The one or more non-transitory storage media of claim 17, wherein performing the operation further comprises:

generating journaling metadata that indicates that transactions are to store change information, caused by the transactions, in a journal;

after generating the journaling metadata, generating cooperation metadata that indicates that subsequent transactions are to check the journal and apply changes, caused by the subsequent transactions, to the new partition.

19. The one or more non-transitory storage media of claim 18, wherein the instructions, when executed by the one or more processors, further cause:

in response to determining a particular time after the journaling metadata is generated and before the cooperation metadata is generated, waiting for one or more transactions that were pending at the particular time to commit before copying the data.

20. The one or more non-transitory storage media of claim 18, wherein the instructions, when executed by the one or more processors, further cause:

after generating the cooperation metadata, identifying a particular change in the journal and applying the particular change to the new partition.

21. The one or more non-transitory storage media of claim 20, wherein:

identifying and applying are performed by a process that is executing a data definition language (DDL) transaction that began prior to creating the new partition;

the process performs copying the data, generating the journaling data, and generating the cooperation metadata.

22. The one or more non-transitory storage media of claim 18, wherein the instructions, when executed by the one or more processors, further cause:

in response to determining a particular time that the journal does not contain any change entries that have not been applied to the new partition, waiting for one or more transactions that were pending at the particular time to commit before exchanging identities of the partition and the new partition.

23. The one or more non-transitory storage media of claim 18, wherein the instructions, when executed by the one or more processors, further cause:

after generating the cooperation metadata, exchanging identities of the partition and the new partition.

24. The one or more non-transitory storage media of claim 18, wherein the instructions, when executed by the one or more processors, further cause:

invalidating one or more first cursors after creating the journal and invalidating one or more second cursors after generating the cooperation metadata.

25. The one or more non-transitory storage media of claim 17, wherein performing the operation comprises:

determining whether a local index exists for the partition;

in response to determining that a local index exists for the partition, generating a new local index for the new partition.

26. The one or more non-transitory storage media of claim 17, wherein the instructions, when executed by the one or more processors, further cause:

generating a plurality of mappings, each of which maps a location identifier for the partition with a location identifier for the new partition.

27. The one or more non-transitory storage media of claim 17, wherein the instructions, when executed by the one or more processors, further cause:

determining that a global index exists for the partition;

in response to receiving a query;

identifying an entry in the global index,

identifying a first location identifier that is indicated in the entry,

based on a mapping in a plurality of mappings, identifying a second location identifier that corresponds to the first location identifier,

using the second location identifier to retrieve data from the new partition.