Hybrid push/pull execution of continuous SQL queries

• US 8,392,402 B2
• Filed: 12/03/2008
• Issued: 03/05/2013
• Est. Priority Date: 12/03/2008
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Mihaila, George Andrei, Stanoi, Ioana Roxana
• Primary Examiner(s): Wilson, Kimberly
• Assistant Examiner(s): Uddin, Mohammed R
• Application Number:
  US12/327,214
• Publication Number:
  US 20100138405A1
• Time in Patent Office: 1,553 Days
• Field of Search: 707/718, 707/764, 707/999.003
• US Class Current: 707/718
• CPC Class Codes: G06F 16/24568 Data stream processing; Continuous queries

Abstract

Illustrative embodiments provide a computer-implemented method for hybrid push/pull of continuous structured query language queries. The computer-implemented method receives stream input, wherein the stream input comprises events of interest, builds and a state machine and stream plans, based on an original query, and replicates the stream input. Responsive to a push sub-query trigger, the computer-implemented method submits a pull sub-query to the database to produce a result, and sends the result to a requester.

Claims

1. A computer-implemented method for hybrid push/pull of continuous structured query language queries, the computer-implemented method comprising:

receiving stream input, wherein the stream input comprises events of interest;

building stream plans based on an original query from the stream input and a state machine, wherein building stream plans comprises splitting the original query into a push sub-query and a pull sub-query, and wherein the pull sub-query represents a portion of the original query from the stream input and the push sub-query represents a remaining portion of the original query from the stream input;

replicating the stream input, wherein replicating the stream input comprises copying the stream input to produce a set of input streams, wherein one of the set of input streams is directed to a database and one of the set of input streams is directed to the state machine;

responsive to a push sub-query trigger, determining whether a pattern of the push sub-query matches an event in the one of the set of input streams directed to the state machine to form a pattern match and creating the push sub-query trigger in response to the pattern match, wherein the pattern match causes submitting of the pull sub-query to execute against accumulated data in the database to produce a result; and

sending the result to a requestor.

2. The computer-implemented method of claim 1, wherein building stream plans further comprises:

creating a set of pull sub-queries, representing a portion of the original query; and

combining the push sub-query and the set of pull sub-queries to create a hybrid push/pull query, wherein the hybrid push/pull query is semantically equivalent to the original query.

3. The computer-implemented method of claim 1, wherein submitting a pull sub-query to the database to produce a result further comprises:

submitting a set of pull sub-queries.

4. The computer-implemented method of claim 1, wherein building stream plans based on an original query further comprises:

decomposing the original query according to selectivities, wherein each of the selectivities represents a percentage of events in the stream input that satisfies a condition; and

evaluating costs of stream plans according to the selectivities.

5. The computer-implemented method of claim 1, wherein the original query is a continuous query.

6. A data processing system for hybrid push/pull of continuous structured query language queries, the data processing system comprising:

a bus;

a memory connected to the bus, wherein the memory comprises computer executable instructions;

a communication unit connected to the bus;

a database connected to the bus;

a processor unit connected to the bus, wherein the processor unit executes the computer executable instructions to direct the data processing system to;

receive stream input, wherein the stream input comprises events of interest;

build stream plans based on an original query from the stream input and a state machine, wherein building stream plans comprises splitting the original query into a push sub-query and a pull sub-query, and wherein the pull sub-query represents a portion of the original query from the stream input and the push sub-query represents a remaining portion of the original query from the stream input;

replicate the stream input, wherein replicating the stream input comprises copying the stream input to produce a set of input streams, wherein one of the set of input streams is directed to a database and one of the set of input streams is directed to the state machine;

responsive to a push sub-query trigger, determine whether a pattern of the push sub-query matches an event in the one of the set of input streams directed to the state machine to form a pattern match and create the push sub-query trigger in response to the pattern match, wherein the pattern match causes submit of the pull sub-query to execute against accumulated data in the database to produce a result; and

send the result to a requestor.

7. The data processing system of claim 6, wherein executing the computer executable instructions to direct the data processing system to build stream plans further comprises executing the computer executable instructions to direct the data processing system to:

create a set of pull sub-queries, representing a portion of the original query; and

combine the push sub-query and the set of pull sub-queries to create a hybrid push/pull query, wherein the hybrid push/pull query is semantically equivalent to the original query.

8. The data processing system of claim 6, wherein executing the computer executable instructions to direct the data processing system to submit a pull sub-query to the database to produce a result further comprises executing the computer executable instructions to direct the data processing system to:

submit a set of pull sub-queries.

9. The data processing system of claim 6, wherein executing the computer executable instructions to direct the data processing system to build stream plans based on an original query further comprises executing the computer executable instructions to direct the data processing system to:

decompose the original query according to selectivities, wherein each of the selectivities represents a percentage of events in the stream input that satisfies a condition; and

evaluate costs of stream plans according to the selectivities.

10. The data processing system of claim 6, wherein the original query is a continuous query.

11. A computer-program product for hybrid push/pull of continuous structured query language queries, the computer program product comprising:

a computer usable storage device, comprising a memory, having computer executable instructions stored thereon, the computer executable instructions comprising;

computer executable instructions for receiving stream input, wherein the stream input comprises events of interest;

computer executable instructions for building stream plans based on an original query from the stream input and a state machine, wherein building stream plans comprises splitting the original query into a push sub-query and a pull sub-query, and wherein the pull sub- query represents a portion of the original query from the stream input and the push sub-query represents a remaining portion of the original query from the stream input;

computer executable instructions for replicating the stream input, wherein replicating the stream input comprises copying the stream input to produce a set of input streams, wherein one of the set of input streams is directed to a database and one of the set of input streams is directed to the state machine;

computer executable instructions, responsive to a push sub-query trigger, for determining whether a pattern of the push sub-query matches an event in the one of the set of input streams directed to the state machine to form a pattern match and creating the push sub-query trigger in response to the pattern match, wherein the pattern match causes submitting of the pull sub-query to execute against accumulated data in the database to produce a result; and

computer executable instructions for sending the result to a requestor.

12. The computer program product of claim 11, wherein computer executable instructions for building stream plans further comprises:

computer executable instructions for creating a set of pull sub-queries, representing a portion of the original query; and

computer executable instructions for combining the push sub-query and the set of pull sub-queries to create a hybrid push/pull query, wherein the hybrid push/pull query is semantically equivalent to the original query.

13. The computer program product of claim 11, wherein computer executable instructions for submitting a pull sub-query to the database to produce a result further comprises:

computer executable instructions for submitting a set of pull sub-queries.

14. The computer program product of claim 11, wherein computer executable instructions for building stream plans based on an original query further comprises:

computer executable instructions for decomposing the original query according to selectivities, wherein the original query is a continuous query, and wherein each of the selectivities represents a percentage of events in the stream input that satisfies a condition; and

computer executable instructions for evaluating costs of stream plans according to the selectivities.