Variable sized database dictionary block encoding

• US 10,558,495 B2
• Filed: 11/25/2014
• Issued: 02/11/2020
• Est. Priority Date: 11/25/2014
• Status: Active Grant

• Current Assignee: SAP SE
• Original Assignee: SAP SE
• Inventors: Schreter, Ivan
• Primary Examiner(s): Gurmu, Muluemebet
• Application Number:
  US14/553,606
• Publication Number:
  US 20160147820A1
• Time in Patent Office: 1,904 Days
• Field of Search: 707693
• CPC Class Codes: G06F 16/2282 Tablespace storage structures; Management thereof
• CPC Class Codes: G06F 9/5016 the resource being the memory

Abstract

Dictionary encoding in a table of a database system is initiated using a single page chain. The database system includes a plurality of processor cores and each page chain includes a plurality of chained pages. Thereafter, n additional page chains are generated for use by the dictionary encoding when the count of pages used by the dictionary encoding reaches a pre-determined limit. Generation of additional page chains is later ceased once the number of additional page chains n is equivalent to a number of available processor cores. Related apparatus, systems, techniques and articles are also described.

Claims

1. A method comprising:

initiating dictionary encoding in a table of a database system using a first page chain, the database system comprising a plurality of processor cores, each page chain comprising a plurality of chained pages;

generating n additional page chains for use by the dictionary encoding when the count of pages used by the dictionary encoding reaches a pre-determined limit;

ceasing generation of further additional page chains once the number of additional page chains n is equivalent to one less than a number of available processor cores; and

storing a new variable sized value to the dictionary, the storing comprising picking a page chain from a plurality of existing page chains corresponding to a current core index of a processor core of the plurality of processor cores on which an executing thread currently runs modulo a number of the plurality of existing page chains, the plurality of existing page chains comprising the first page chain and then additional page chains, the current core index corresponding to the processor core on which the executing thread is currently running.

2. The method of claim 1, wherein the page chains comprise a set of active pages on which write operations are accepted.

3. The method of claim 2, wherein the database comprises an in-memory architecture.

4. The method of claim 3, wherein the active pages are held in memory.

5. The method of claim 1, wherein the database comprises a non-uniform memory access (NUMA) architecture.

6. The method of claim 5, wherein the NUMA architecture comprises a plurality of NUMA nodes, wherein processors forming part of the same NUMA node are assigned to the same page chain.

7. A non-transitory computer program product storing instructions which, when executed by at least one hardware data processor forming part of at least one computing system, result in operations comprising:

initiating dictionary encoding in a table of a database system using a single page chain, the database system comprising a plurality of processor cores, each page chain comprising a plurality of chained pages;

generating n additional page chains for use by the dictionary encoding when the count of pages used by the dictionary encoding reaches a pre-determined limit;

ceasing generation of further additional page chains once the number of additional page chains n is equivalent to one less than a number of available processor cores; and

storing a new variable sized value to the dictionary, the storing comprising picking a page chain from a plurality of existing page chains corresponding to a current core index of a processor core of the plurality of processor cores on which an executing thread currently runs modulo a number of the plurality of existing page chains, the plurality of existing page chains comprising the first page chain and then additional page chains, the current core index corresponding to the processor core on which the executing thread is currently running.

8. The computer program product of claim 7, wherein the page chains comprise a set of active pages on which write operations are accepted.

9. The computer program product of claim 8, wherein the database comprises an in-memory architecture.

10. The computer program product of claim 9, wherein the active pages are held in memory.

11. The computer program product of claim 7, wherein the database comprises a non-uniform memory access (NUMA) architecture.

12. The computer program product of claim 11, wherein the NUMA architecture comprises a plurality of NUMA nodes, wherein processors forming part of the same NUMA node are assigned to the same page chain.

13. A system comprising:

at least one data processor; and

memory storing instructions which, when executed by the at least one data processor, result in operations comprising;

initiating dictionary encoding in a table of a database system using a single page chain, the database system comprising a plurality of processor cores, each page chain comprising a plurality of chained pages;

generating n additional page chains for use by the dictionary encoding when the count of pages used by the dictionary encoding reaches a pre-determined limit; and

ceasing generation of further additional page chains once the number of additional page chains n is equivalent to one less than a number of available processor cores; and

storing a new variable sized value to the dictionary, the storing comprising picking a page chain from a plurality of existing page chains corresponding to a current core index of a processor core of the plurality of processor cores on which an executing thread currently runs modulo a number of the plurality of existing page chains, the plurality of existing page chains comprising the first page chain and the n additional page chains, the current core index corresponding to the processor core on which the executing thread is currently running.

14. The system of claim 13, wherein the page chains comprise a set of active pages on which write operations are accepted.

15. The system of claim 13, wherein the at least one processor and memory form part of the database and the database is an in-memory database having a non-uniform memory access (NUMA) architecture that comprises a plurality of NUMA nodes, wherein processors forming part of the same NUMA node are assigned to the same page chain.