Method and system for parallelizing database requests

• US 20110202929A1
• Filed: 07/07/2010
• Published: 08/18/2011
• Est. Priority Date: 02/16/2010
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Schleimer, Stephen, Rishel, Ryder B., Taylor, Derek A.
• Granted Patent: US 9,171,044 B2
• Time in Patent Office: Days
• US Class Current: 718/106
• CPC Class Codes: G06F 16/24549 Run-time optimisation

Abstract

Methods and systems are described for applying the use of shards within a single memory address space. A database request is processed by providing the request from a client to a processor, the processor then distributing the request to multiple threads within a single process but executing in a shared memory address environment, wherein each thread performs the request on a distinct shard, and aggregating the results of the multiple threads being aggregated and returning a final result to the client. By parallelizing operations in this way, the request response time can be reduced and the total amount of communication overhead can be reduced.

Claims

1. A method of executing a database request using a server comprising a plurality of processing resources, the method comprising:

(a) receiving the database request;

(b) transforming the database request into a plurality of sub-requests; and

(c) executing the different ones of the plurality of sub-requests on different ones of the plurality of processing resources, each of the plurality of processing resources operating on respective autonomous data subsets by performing operations without consideration for data in other data subsets;

wherein the respective autonomous data subsets occupy a common address space.

2. The method of claim 1, wherein the common address space is a common virtual address space.

3. The method of claim 1, wherein the sub-requests of part (b) are similar sub-requests.

4. The method of claim 1, wherein each processing resource runs in a processing thread running on a processor core.

5. The method of claim 1, wherein the results of the sub-requests of (c) are aggregated.

6. The method of claim 5, wherein the receiving database request comes from a client, and wherein the aggregated results are returned to the client.

7. A computer-readable medium comprising instructions for performing the steps of:

providing access to a database distributed into non-overlapping subsets within a memory,

directing a plurality of threads, running on a plurality of cores within a common address space, to conduct a request of the database, wherein each thread performs a sub-request in relation a unique subset of the database, and

combining results generated by each thread.

8. The computer readable medium of claim 7, wherein the common address space is a common virtual address space.

9. The computer readable medium of claim 7, wherein said sub-requests are similar sub-requests.

10. The computer readable medium of claim 7, comprising instructions for causing each thread to run on its own core.

11. The computer readable medium of claim 7, comprising instructions for aggregating results of the sub-requests to produce aggregated results.

12. The computer readable medium of claim 11, comprising instructions for receiving a database request from a client, and for returning the aggregated results to the client.

13. A system for responding to a database request comprising:

a memory for storing database records within a common memory space,

a plurality of processing resources residing on a single server and configured to execute a plurality of processing threads in a time-overlapping manner, and

a database divided into multiple, non-overlapping subsets within the common memory space;

wherein the processing resources are configured to process a database request such that each of the plurality of threads performs a sub-request with respect to a different subset of the database.

14. The system of claim 13, wherein the combination of the database subsets contain a complete set of database records, indexes, and transactional consistency states.

15. The system of claim 14, wherein each of the database subsets contains its own set of database records, indexes and transactional consistency states.

16. The system of claim 14, wherein the processing resources are configured such that each of the plurality of threads performs a sub-request with respect to a different subset of the database including indexes and transactional consistency states pertaining to that subset of the database.

17. A database server comprising:

a memory subsystem for storing a plurality of database micro-shards;

a client input queue for receiving a database request from a client and for producing a plurality of sub-requests for a corresponding plurality of micro-shards in response to the database request;

a plurality of processor cores configured to execute the plurality of sub-requests in a time-overlapping manner; and

an aggregation processor for receiving results of execution of the plurality of sub-requests and aggregating the results to produce a final result to be communicated to the client.

18. The database server of claim 17, wherein the database micro-shards share a common address space.

19. The database server of claim 18, wherein the common address space is a virtual address space.

20. The database server of claim 17, wherein the memory subsystem is a semiconductor memory subsystem.