Database table re-partitioning using two active partition specifications

Patent metadata

Field	Value
Publication number	US-10042910-B2
Application number	US-201414188541-A
Country	US
Kind code	B2
Filing date	Feb 24, 2014
Priority date	Feb 24, 2014
Publication date	Aug 7, 2018
Grant date	Aug 7, 2018

Abstract

Partitioning of source partitions of a table of a database to target partitions is initiated. Thereafter, a transition partition specification is specified that identifies the source partitions and the target partitions. Data is then moved (e.g., asynchronously moved, etc.) from the source partitions to the target partitions. Concurrently with the moving of the data, operates are handled using the transition partition specification. Subsequently, the source partitions are dropped when all of the data has been moved to the target partitions and there are no open transactions accessing the source partitions. Related apparatus, systems, techniques and articles are also described.

First claim

What is claimed is:

1. A computer-implemented method, comprising:
initiating a repartitioning of a table of a database, the table being repartitioned from source partitions to target partitions;
specifying a transition partition specification identifying the source partitions and the target partitions;
receiving, from a client, an operation with respect to the table; and
concurrently executing the operation and moving data from the source partitions to the target partitions, the moving of the data being based at least on the transition partition specification, the data being moved on a row level by at least starting a data move transaction as a part of a background job, selecting data from a source partition, inserting the selected data into a target partition using an insert stack forming part of the database, deleting the selected data from the source partition, and committing the data move transaction, the operation being executed concurrently with the moving of the data by at least:
in response to the operation deleting at least one row of data from the table, executing the operation by at least deleting the at least one row of data from both the source partitions and the target partitions;
in response to the operation updating the at least one row of data from the table, executing the operation by at least:
deleting the at least one row of data from both the source partitions and the target partitions; and
inserting at least one updated row of data into the target partitions but not the source partitions; and
in response to the operation inserting the at least one row of data into the table, executing the operation by at least inserting the at least one row of data into the target partitions but not the source partitions;
dropping the source partitions when all of the data has been moved to the target partitions and there are no open transactions accessing the source partitions; and
exclusively locking the table and updating the transition partition specification to reflect the data as having been moved to the target partitions.

2. The method as in claim 1 , wherein the data is moved on a column level.

3. The method as in claim 2 , wherein the data is moved on the column level using a remote agent.

4. The method as in claim 1 , wherein the database is an in-memory column-oriented database.

5. A non-transitory computer program product storing instructions which, when implemented by at least one data processor forming part of at least one computing system, result in operations comprising:
initiating a repartitioning of a table of a database, the table being repartitioned from source partitions to target partitions;
specifying a transition partition specification identifying the source partitions and the target partitions;
receiving, from a client, an operation with respect to the table; and
concurrently executing the operation and moving data from the source partitions to the target partitions, the moving of the data being based at least on the transition partition specification, the data being moved on a row level by at least starting a data move transaction as a part of a background job, selecting data from a source partition, inserting the selected data into a target partition using an insert stack forming part of the database, deleting the selected data from the source partition, and committing the data move transaction, the operation being executed concurrently with the moving of the data by at least:
in response to the operation deleting at least one row of data from the table, executing the operation by at least deleting the at least one row of data from both the source partitions and the target partitions;
in response to the operation updating the at least one row of data from the table, executing the operation by at least:
deleting the at least one row of data from both the source partitions and the target partitions; and
inserting at least one updated row of data into the target partitions but not the source partitions; and
in response to the operation inserting the at least one row of data into the table, executing the operation by at least inserting the at least one row of data into the target partitions but not the source partitions;
dropping the source partitions when all of the data has been moved to the target partitions and there are no open transactions accessing the source partitions; and
exclusively locking the table and updating the transition partition specification to reflect the data as having been moved to the target partitions.

6. The computer program product as in claim 5 , wherein the data is moved on a column level.

7. The computer program product as in claim 6 , wherein the data is moved on the column level using a remote agent.

8. The computer program product as in claim 5 , wherein the database is an in-memory column-oriented database.

9. A system comprising:
at least one data processor; and
memory storing instructions which, when implemented by the at least one data processor, result in operations comprising:
initiating a repartitioning of a table of a database, the table being repartitioned from source partitions to target partitions;
specifying a transition partition specification identifying the source partitions and the target partitions;
receiving, from a client, an operation with respect to the table; and
concurrently executing the operation and moving data from the source partitions to the target partitions, the moving of the data being based at least on the transition partition specification, the data being moved on a row level by at least starting a data move transaction as a part of a background job, selecting data from a source partition, inserting the selected data into a target partition using an insert stack forming part of the database, deleting the selected data from the source partition, and committing the data move transaction, the operation being executed concurrently with the moving of the data by at least:
in response to the operation deleting at least one row of data from the table, executing the operation by at least deleting the at least one row of data from both the source partitions and the target partitions;
in response to the operation updating the at least one row of data from the table, executing the operation by at least:
deleting the at least one row of data from both the source partitions and the target partitions; and
inserting at least one updated row of data into the target partitions but not the source partitions; and

in response to the operation inserting the at least one row of data into the table, executing the operation by at least inserting the at least one row of data into the target partitions but not the source partitions;
dropping the source partitions when all of the data has been moved to the target partitions and there are no open transactions accessing the source partitions; and
exclusively locking the table and updating the transition partition specification to reflect the data as having been moved to the target partitions.

10. The system as in claim 9 , wherein the data is moved on a column level using a remote agent.

11. The system as in claim 9 , wherein the database is an in-memory column-oriented database.

Assignees

• Bensberg Christian
• Fricke Lars
• Lee Juchang
• Muehle Michael
• Peh Thomas
• Stephan Wolfgang
• Sap Se

Inventors

• Bensberg Christian
• Fricke Lars
• Lee Juchang
• Muehle Michael
• Peh Thomas
• Stephan Wolfgang

Classifications

• G06F16/273

  Asynchronous replication or reconciliation · CPC title

• G06F16/278 Primary

  Data partitioning, e.g. horizontal or vertical partitioning · CPC title

• G06F17/30584 Primary

  Physics · mapped topic

• G06F17/30578

  Physics · mapped topic