Method for semantic modeling of stream processing components to enable automatic application composition

• US 8,098,248 B2
• Filed: 04/02/2007
• Issued: 01/17/2012
• Est. Priority Date: 04/02/2007
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Feblowitz, Mark D., Liu, Zhen, Ranganathan, Anand, Riabov, Anton V.
• Primary Examiner(s): Amini, Javid
• Application Number:
  US11/695,457
• Publication Number:
  US 20080243451A1
• Time in Patent Office: 1,751 Days
• Field of Search: 345/440
• US Class Current: 345/440
• CPC Class Codes: G06F 16/24568 Data stream processing; Continuous queries

Abstract

A method for modeling components of a stream processing application, includes: defining an input message pattern of a processing element, wherein the input message pattern includes variables representing data objects that must be included in a message input to the processing element, and a graph pattern that semantically describes the data objects that must be included in the message input to the processing element; and defining an output message pattern of the processing element, wherein the output message pattern includes variables and new data objects, the variables and new data objects representing data objects that must be included in a message output from the processing element, and a graph pattern that semantically describes the data objects in the message output from the processing element.

Claims

1. A method for modeling components of a stream processing application, comprising:

defining an input message pattern of a processing element, wherein the input message pattern includes variables representing data objects that must be included in a message input to the processing element, and a graph pattern that semantically describes the data objects that must be included in the message input to the processing element; and

defining an output message pattern of the processing element, wherein the output message pattern includes variables and new data objects, the variables and new data objects representing data objects that must be included in a message output from the processing element, and a graph pattern that semantically describes the data objects in the message output from the processing element,

wherein the method is performed using a processor and every one of the variables in the output message pattern must be included in the input message pattern.

2. The method of claim 1, wherein the graph pattern that semantically describes the data objects that must be included in the message input to the processing element uses terms defined in ontologies represented in Web Ontology Language (OWL).

3. The method of claim 1, wherein the graph pattern that semantically describes the data objects in the message output from the processing element uses terms defined in ontologies represent in OWL.

4. The method of claim 1, further comprising:

defining a set of variables that represent configuration parameters of the processing element.

5. The method of claim 4, further comprising:

inputting a plurality of modeled processing element descriptions, each processing element description having a defined input and output message pattern and defined configuration;

inputting a processing request, wherein the processing request includes a goal that is represented by a graph pattern that semantically describes a desired processing outcome; and

assembling a processing graph that produces a stream that satisfies the desired processing outcome.

6. The method of claim 5, wherein assembling the processing graph comprises:

matching a message output from a first processing element to an input message pattern of a second processing element if the message includes the data objects that must be included in a message input to the second processing element and if a graph that semantically describes data objects in the message satisfies the graph pattern that semantically describes the data objects that must be included in the message input to the second processing element.

7. The method of claim 6, further comprising:

connecting the first processing element to the second processing element when the message is matched to the input message pattern of the second processing element.

8. The method of claim 7, further comprising:

generating an output message of the second processing element by applying differences between the graph pattern that semantically describes the data objects that must be included in the message input to the second processing element and the graph pattern that semantically describes the data objects produced as output by the second processing element to the graph that semantically describes the data objects in the message based on a graph transformation operation.

9. The method of claim 1, further comprising:

defining an output message pattern of a data source, wherein the output message includes exemplar terms, the exemplar terms representing data objects that must be included in a message output from the data source, and a graph that semantically describes the data objects in the message output from the data source.

10. The method of claim 9, wherein the graph that semantically describes the data objects in the message output from the data source is formed by OWL assertions.

11. The method of claim 9, further comprising:

inputting a plurality of modeled processing element descriptions, each processing element description having a defined input and output message pattern;

inputting a plurality of data source descriptions, each data source description having a defined output message pattern;

inputting a processing request, wherein the processing request includes a goal that is represented by a graph pattern that semantically describes a desired processing outcome; and

assembling a processing graph that includes at least one of the modeled processing elements or at least one of the modeled data sources that produces output messages that satisfy the desired processing outcome.

12. The method of claim 11, wherein assembling the processing graph comprises:

matching a message output from a data source to an input message pattern of a processing element if the message includes the data objects that must be included in a message input to the processing element and if a graph that semantically describes data objects in the message satisfies the graph pattern that semantically describes the data objects that must be included in the message input to the processing element.

13. The method of claim 12, further comprising:

connecting the data source to the processing element when the message is matched to the input message pattern of the processing element.

14. The method of claim 13, further comprising:

generating an output message of the processing element by applying differences between the graph that semantically describes data objects in the message and the graph pattern that semantically describes the data objects that must be included in the message input to the processing element to the graph pattern that semantically describes the data objects in the message output from the processing element based on a graph transformation operation.

15. A system for modeling components of a stream processing application, comprising:

a memory device for storing a program;

a processor in communication with the memory device, the processor operative with the program to;

define an input message pattern of a processing element, wherein the input message pattern includes variables representing data objects that must be included in a message input to the processing element, and a graph pattern that semantically describes the data objects that must be included in the message input to the processing element; and

define an output message pattern of the processing element, wherein the output message pattern includes variables and new data objects, the variables and new data objects representing data objects that must be included in a message output from the processing element, and a graph pattern that semantically describes the data objects in the message output from the processing element,

wherein every one of the variables in the output message pattern must be included in the input message pattern.

16. The system of claim 15, wherein the graph pattern that semantically describes the data objects that must be included in the message input to the processing element uses terms defined in ontologies represented in Web Ontology Language (OWL).

17. The system of claim 15, wherein the graph pattern that semantically describes the data objects in the message output from the processing element uses terms defined in ontologies represent in OWL.

18. The system of claim 15, wherein the processor is further operative with the program to:

define a set of variables that represent configuration parameters of the processing element.

19. The system of claim 18, wherein the processor is further operative with the program to:

receive a plurality of modeled processing element descriptions, each processing element description having a defined input and output message pattern and defined configuration;

receive a processing request, wherein the processing request includes a goal that is represented by a graph pattern that semantically describes a desired processing outcome; and

assemble a processing graph that produces a stream that satisfies the desired processing outcome.

20. The system of claim 19, wherein when assembling the processing graph the processor is further operative with the program to:

match a message output from a first processing element to an input message pattern of a second processing element if the message includes the data objects that must be included in a message input to the second processing element and if a graph that semantically describes data objects in the message satisfies the graph pattern that semantically describes the data objects that must be included in the message input to the second processing element.

21. The system of claim 20, wherein the processor is further operative with the program to:

connect the first processing element to the second processing element when the message is matched to the input message pattern of the second processing element.

22. The system of claim 21, wherein the processor is further operative with the program to:

generate an output message of the second processing element by applying differences between the graph pattern that semantically describes the data objects that must be included in the message input to the second processing element and the graph pattern that semantically describes the data objects produced as output by the second processing element to the graph that semantically describes the data objects in the message based on a graph transformation operation.

23. The system of claim 15, wherein the processor is further operative with the program to:

define an output message pattern of a data source, wherein the output message includes exemplar terms, the exemplar terms representing data objects that must be included in a message output from the data source, and a graph that semantically describes the data objects in the message output from the data source.

24. The system of claim 23, wherein the graph that semantically describes the data objects in the message output from the data source is formed by OWL assertions.

25. The system of claim 23, wherein the processor is further operative with the program to:

receive a plurality of modeled processing element descriptions, each processing element description having a defined input and output message pattern;

receive a plurality of data source descriptions, each data source description having a defined output message pattern;

receive a processing request, wherein the processing request includes a goal that is represented by a graph pattern that semantically describes a desired processing outcome; and

assemble a processing graph that includes at least one of the modeled processing elements or at least one of the modeled data sources that produces output messages that satisfy the desired processing outcome.

26. The system of claim 25, wherein when assembling the processing graph the processor is further operative with the program to:

match a message output from a data source to an input message pattern of a processing element if the message includes the data objects that must be included in a message input to the processing element and if a graph that semantically describes data objects in the message satisfies the graph pattern that semantically describes the data objects that must be included in the message input to the processing element.

27. The system of claim 26, wherein the processor is further operative with the program to:

connect the data source to the processing element when the message is matched to the input message pattern of the processing element.

28. The system of claim 27, wherein the processor is further operative with the program to:

generate an output message of the processing element by applying differences between the graph that semantically describes data objects in the message and the graph pattern that semantically describes the data objects that must be included in the message input to the processing element to the graph pattern that semantically describes the data objects in the message output from the processing element based on a graph transformation operation.

29. A computer program product comprising a non-transitory computer useable medium having computer program logic recorded thereon for modeling components of a stream processing application, the computer program logic comprising:

program code for defining an input message pattern of a processing element, wherein the input message pattern includes variables representing data objects that must be included in a message input to the processing element, and a graph pattern that semantically describes the data objects that must be included in the message input to the processing element; and

program code for defining an output message pattern of the processing element, wherein the output message pattern includes variables and new data objects, the variables and new data objects representing data objects that must be included in a message output from the processing element, and a graph pattern that semantically describes the data objects in the message output from the processing element,

wherein every one of the variables in the output message pattern must be included in the input message pattern.