Reassignment of data to different configurations of processing units in database systems

• US 10,459,907 B2
• Filed: 10/30/2017
• Issued: 10/29/2019
• Est. Priority Date: 12/08/2014
• Status: Active Grant

• Current Assignee: Teradata US, Inc. (Teradata Corporation)
• Original Assignee: Teradata US, Inc. (Teradata Corporation)
• Inventors: Kaufmann, Frederick S., Kolandasamy, Rajendran, Benton, Philip Jason, Sinclair, Paul Laurence, Pederson, Donald Raymond, Burger, Louis Martin
• Primary Examiner(s): Vy, Hung T
• Application Number:
  US15/797,188
• Publication Number:
  US 20180052877A1
• Time in Patent Office: 729 Days
• Field of Search: 707700, 707791, 707713
• CPC Class Codes: G06F 16/2282 Tablespace storage structures; Management thereof
• CPC Class Codes: G06F 16/24532 of parallel queries
• CPC Class Codes: G06F 16/24542 Plan optimisation
• CPC Class Codes: G06F 16/24569 Query processing with adaptation to specific hardware, e.g. adapted for using GPUs or SSDs

Abstract

An order (or priority) for reassigning data portions of a database from a first map to a second map can be determined. It should be noted that the first map assigns the data portions to a first configuration of processors for processing the data portions, and the second map assigns the data portions to a second configuration of processors, different than the first configuration, for processing the data portions in a database system and/or environment. The order can be determined by grouping the data in multiple groups and ordering the groups in accordance with a determined priority for reassignment. In doing so, one or more relationships between the data portions and/or cost of use for each one the groups can be considered.

Claims

1. A computer-implemented method of reassigning data portions of a database from a first map to a second map for processing by multiple processing units of a database system configured to at least process data stored in the database, wherein the computer-implemented method is implemented by one or more physical processors configured to at least process data of the database stored in a non-transitory computer storage medium, and wherein the computer-implemented method comprises:

obtaining data portions assigned to the first map in an initial order for reassignment to the second map wherein the first map assigns the data portions to a first configuration of processors for processing the data portions, and the second map assigns the data portions to a second configuration of processors, different than the first configuration, for processing the data portions;

determining, based on one or more relationships between the data portions in the initial order, multiple groups for reassignment from the first map to the second map, wherein each one of the multiple groups includes two or more of the data portions of the initial ordered data portions;

determining a cost of use for each one of the multiple groups for reassignment from the first map to the second map;

determining, based on the cost of the use of each one of the multiple groups, a second order of reassignment of the multiple groups from the first map to the second map; and

processing, by said second configuration of processors, said data portions in accordance with said second map.

2. The computer-implemented method of claim 1, wherein the determining of the multiple groups based on one or more relationships between the data portions in the initial order for reassignment further comprises:

determining that the data portions need to be joined on the same one of the processors.

3. The computer-implemented method of claim 1, wherein determining the cost of use for each one of the multiple groups for reassignment from the first map to the second map further comprises:

determining at least one of;

(a) frequency of use and (b) cost of each use of each one of the data portions of the multiple groups.

4. The computer-implemented method of claim 1, wherein determining based on the cost of the use of each one of the multiple groups a second order of reassignment of the multiple groups from the first map to the second map further comprises:

prioritizing the multiple groups such that a first group of the multiple groups that has a higher determined cost than a second group of the multiple groups is prioritized to be reassigned from the first map to the second map before the second group of the multiple groups.

5. The computer-implemented method of claim 1, wherein the computer-implemented method further comprises:

arranging the data portions assigned to the first map in accordance with their size to obtain the initial order for reassignment to the second map.

6. The computer-implemented method of claim 1, wherein the computer-implemented method further comprises:

removing duplicate groups.

7. The computer-implemented method of claim 1, wherein the data portions are distinct database tables of the database.

8. A computing system that includes one or more physical processors configured to at least process data of the database stored in a non-transitory computer storage medium, wherein the one or more physical processors are configured to:

obtain data portions assigned to the first map in an initial order for reassignment to a second map, wherein the first map assigns the data portions to a first configuration of processors for processing the data portions, and the second map assigns the data portions to a second configuration of processors, different than the first configuration, for processing the data portions;

determine, based on one or more relationships between the data portions in the initial order, multiple groups for reassignment from the first map to the second map, wherein each one of the multiple groups includes two or more of the data portions of the initial ordered data portions;

determine a cost of use for each one of the multiple groups for reassignment from the first map to the second map;

determine, based on the cost of the use of each one of the multiple groups, a second order of reassignment of the multiple groups from the first map to the second map; and

process said data portions in accordance with said second map.

9. The computing system as recited in claim 8, wherein the determining of the multiple groups based on one or more relationships between the data portions in the initial order for reassignment further comprises:

determining that the data portions need to be joined on the same one of the processors.

10. The computing system as recited in claim 8, wherein determining the cost of use for each one of the multiple groups for reassignment from the first map to the second map further comprises:

determining at least one of;

(a) frequency of use and (b) cost of each use of each one of the data portions of the multiple groups.

11. The computing system as recited in claim 8, wherein determining based on the cost of the use of each one of the multiple groups a second order of reassignment of the multiple groups from the first map to the second map further comprises:

prioritizing the multiple groups such that a first group of the multiple groups that has a higher determined cost than a second group of the multiple groups is prioritized to be reassigned from the first map to the second map before the second group of the multiple groups.

12. The computing system as recited in claim 8, wherein the one or more physical processors are further configured to:

arrange the data portions assigned to the first map in accordance with their size to obtain the initial order for reassignment to the second map.

13. The computing system as recited in claim 8, wherein the one or more physical processors are further configured to:

remove duplicate groups.

14. The computing system as recited in claim 8, wherein the data portions are distinct database tables of the database.

15. A non-transitory computer readable medium storing at least computer executable code that reassigns data portions of a database from a first map to a second map for processing by multiple processing units of a database system configured to at least process data stored in the database, wherein the executable code when executed:

obtains data portions assigned to the first map in an initial order for reassignment to a second map, wherein the first map assigns the data portions to a first configuration of processors for processing the data portions, and the second map assigns the data portions to a second configuration of processors, different than the first configuration, for processing the data portions;

determines, based on one or more relationships between the data portions in the initial order, multiple groups for reassignment from the first map to the second map, wherein each one of the multiple groups includes two or more of the data portions of the initial ordered data portions;

determines a cost of use for each one of the multiple groups for reassignment from the first map to the second map;

determines, based on the cost of the use of each one of the multiple groups, a second order of reassignment of the multiple groups from the first map to the second map; and

processes said data portions in accordance with said second map.

16. The non-transitory computer readable medium as recited in claim 15, wherein the determining of the multiple groups based on one or more relationships between the data portions in the initial order for reassignment further comprises:

determining that the data portions need to be joined on the same one of the processors.

17. The non-transitory computer readable medium as recited in claim 15, wherein determining the cost of use for each one of the multiple groups for reassignment from the first map to the second map further comprises:

determining at least one of;

(a) frequency of use and (b) cost of each use of each one of the data portions of the multiple groups.

18. The non-transitory computer readable medium as recited in claim 15, wherein determining based on the cost of the use of each one of the multiple groups a second order of reassignment of the multiple groups from the first map to the second map further comprises:

prioritizing the multiple groups such that a first group of the multiple groups that has a higher determined cost than a second group of the multiple groups is prioritized to be reassigned from the first map to the second map before the second group of the multiple groups.

19. The non-transitory computer readable medium as recited in claim 15, wherein the executable code when executed further:

arranges the data portions assigned to the first map in accordance with their size to obtain the initial order for reassignment to the second map.

20. The non-transitory computer readable medium as recited in claim 15, wherein the executable code when executed further:

removes duplicate groups.