Redistributing computation work between data producers and data consumers

• US 8,935,227 B2
• Filed: 04/17/2012
• Issued: 01/13/2015
• Est. Priority Date: 04/17/2012
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Zhou, Johnny Quan, Choudhury, Nilesh, Loaiza, Juan R., Potapov, Dmitry Mikhailovich, Tsukerman, Alexander, Umamageswaran, Kothanda
• Primary Examiner(s): Perveen, Rehana
• Assistant Examiner(s): Khong, Alexander
• Application Number:
  US13/449,192
• Publication Number:
  US 20130275402A1
• Time in Patent Office: 1,001 Days
• Field of Search: 707/705
• US Class Current: 707/705
• CPC Class Codes: G06F 16/24552 Database cache management
• CPC Class Codes: G06F 16/24557 Efficient disk access during query execution

Abstract

In a consumer-producer distributed model of computing, consumers request data from producers and perform computing work on the data received from the producers, thereby transforming the data. At least some of the computing work performed by a consumer on data provided by the producer is commutable between the consumer and producer, that is, either the producer as well as the consumer can perform the computing work on the data. A producer redistributes commutable based on the work load of the producer and consumer as well as other factors.

Claims

1. A method, comprising:

a storage device receiving a worked data block request from a database server, said worked data block request requesting worked data blocks generated by commutable work performed by said storage device on storage data blocks, wherein said storage device and said database server are configured to perform said commutable work, wherein said database server is configured to transmit said worked data block request to compute a query conforming to a database language;

based on one or more factors, said storage device determining to push back at least a portion of said worked data blocks requested by said worked data block request as unworked data blocks;

in response to determining to push back at least a portion of said worked data blocks, returning to said database server unworked data blocks in lieu of said portion of said worked data blocks;

wherein said storage device comprises one or more computing nodes; and

wherein the steps of the method are performed by said one or more computing nodes.

2. The method of claim 1, wherein the one or more factors include work load on the storage device.

3. The method of claim 1, wherein the one or more factors include work load on the database server.

4. The method of claim 3, the steps of further including:

the storage device maintaining a flag indicating whether said database server has available resources to accept pushed back commutable work;

the storage device receiving from the database server updates to the flag; and

wherein the one or more factors include the flag.

5. The method of claim 1, wherein the one or more factors include a type of the commutable work.

6. A non-transitory computer-readable medium storing instructions which, when executed by one or more processors, cause:

a storage device receiving a worked data block request from a database server, said worked data block request requesting worked data blocks generated by commutable work performed by said storage device on storage data blocks, wherein said storage device and said database server are configured to perform said commutable work, wherein said database server is configured to transmit said worked data block request to compute a query conforming to a database language;

based on one or more factors, said storage device determining to push back at least a portion of said worked data blocks requested by said worked data block request as unworked data blocks;

in response to determining to push back at least a portion of said worked data blocks, returning to said database server unworked data blocks in lieu of said portion of said worked data blocks; and

wherein said storage device comprises one or more computing nodes.

7. The non-transitory computer-readable medium of claim 6, wherein the one or more factors include work load on the storage device.

8. The non-transitory computer-readable medium of claim 6, wherein the one or more factors include work load on the database server.

9. The non-transitory computer-readable medium of claim 8, the instructions further including instructions for:

the storage device maintaining a flag indicating whether said database server has available resources to accept pushed back commutable work;

the storage device receiving from the database server updates to the flag; and

wherein the one or more factors include the flag.

10. The non-transitory computer-readable medium of claim 6, wherein the one or more factors include a type of the commutable work.