Joining operator graph elements via whole program optimization

• US 10,175,962 B2
• Filed: 07/02/2015
• Issued: 01/08/2019
• Est. Priority Date: 06/10/2015
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Barsness, Eric L., Branson, Michael J., Carey, James E., Santosuosso, John M.
• Primary Examiner(s): Morshed, Hossain M
• Application Number:
  US14/791,046
• Publication Number:
  US 20160364218A1
• Time in Patent Office: 1,286 Days
• Field of Search: 717151, 717154, 717155, 717156, 717157
• CPC Class Codes: G06F 8/41 Compilation
• CPC Class Codes: G06F 8/43 Checking; Contextual analysis
• CPC Class Codes: G06F 9/46 Multiprogramming arrangements
• CPC Class Codes: G06F 9/5027 the resource being a machine, e.g. CPUs, Servers, Terminals

Abstract

Techniques disclosed herein describe joining processing elements (e.g., of a streams processing environment) using optimization techniques, such as by whole program optimization. An optimization tool evaluates multiple processing elements based on one or more optimization rules. The optimization tool identifies, based on the evaluation, processing elements that can be merged. The optimization tool joins the identified processing elements together.

Claims

1. A computer-implemented method of optimizing distributed applications by joining processing elements, the computer-implemented method comprising:

analyzing respective instruction code of each of a plurality of processing elements of a distributed application, wherein the plurality of processing elements are organized in an operator graph, wherein the plurality of processing elements includes at least a first processing element and a second processing element, which are of distinct processing element types and which are not in succession relative to one another in the operator graph;

precompiling the operator graph in order to save compiler artifacts corresponding to the first processing element and the second processing element, the compiler artifacts including instruction code and symbols, wherein the compiler artifacts are evaluated based on one or more specified rules;

determining, by operation of one or more computer processors and based on at least one of the analysis and the evaluation, that the instruction code of the first processing element and the instruction code of the second processing element satisfy at least one optimization rule;

upon determining that the at least one optimization rule is satisfied, joining the instruction code of the first processing element with the instruction code of the second processing element based on the at least one optimization rule and in order to generate a third processing element which includes at least some of the instruction code from both the first processing element and the second processing element; and

deploying the distributed application, which includes the third processing element and excludes both the first processing element and the second processing element, thereby reducing a processing overhead incurred by the distributed application relative to inclusion of both the first processing element and the second processing element.

2. The computer-implemented method of claim 1, further comprising, prior to joining the instruction code of the first processing element with the instruction code of the second processing element:

generating a suggestion to join the instruction code of the first processing element with the instruction code of the second processing element; and

presenting the suggestion in an interface associated with the operator graph.

3. The computer-implemented method of claim 2, further comprising, prior to joining the instruction code of the first processing element with the instruction code of the second processing element:

receiving, via the interface, a selection of the first processing element and the second processing element in response to the suggestion.

4. The computer-implemented method of claim 1, wherein the at least one optimization rule specifies that the instruction code of the first processing element should be joined with the instruction code of the second processing element based on at least one of (i) the type of the first processing element being compatible with the type of the second processing element and (ii) a function performed by the instruction code of the first processing element being compatible with a function performed by the instruction code of the second processing element.

5. The computer-implemented method of claim 1, wherein the plurality of processing elements are of a streams processing environment.

6. The computer-implemented method of claim 1, wherein analyzing the plurality of processing elements comprises:

analyzing the instruction code of the first processing element relative to the instruction code of each of the other plurality of processing elements in the operator graph as the second processing element based on the at least one optimization rule.

7. The computer-implemented method of claim 1, wherein joining the first processing element with the second processing element comprises:

merging the instruction code of the first processing element and the instruction code of the second processing element into the third processing element.

8. The computer-implemented method of claim 7, wherein the distributed application executes in a streams processing environment, the computer-implemented method further comprising:

subsequent to merging the instruction code of the first and second processing elements into the third processing element, removing the first and second processing elements from the operator graph, wherein the distributed application is deployed to the streams processing environment.

9. The computer-implemented method of claim 8, wherein the instruction code of the plurality of processing elements is generated by a compiler, wherein the instruction code of the plurality of processing elements comprises executable instruction code.

10. The computer-implemented method of claim 1, wherein the computer-implemented method is performed by an optimization tool component of an integrated development environment (IDE) tool, wherein the optimization tool component comprises a plurality of subcomponents, the plurality of subcomponents including a retrieval subcomponent, an evaluation subcomponent, an identification subcomponent, and a merge subcomponent.

11. The computer-implemented method of claim 10, wherein the retrieval subcomponent is configured to obtain the respective instruction code of each of the plurality of processing elements;

wherein the evaluation subcomponent is configured to analyze the respective instruction code of each of the plurality of processing elements;

wherein the identification subcomponent is configured to determine that the instruction code of the first processing element and the instruction code of the second processing element satisfy the at least one optimization rule;

wherein the merge subcomponent is configured to automatically join the instruction code of the first processing element with the instruction code of the second processing element in order to generate the third processing element.

12. The computer-implemented method of claim 11, wherein the plurality of processing elements are of a streams processing environment, wherein the computer-implemented method further comprises, prior to joining the instruction code of the first processing element with the instruction code of the second processing element:

generating a suggestion to join the instruction code of the first processing element with the instruction code of the second processing element; and

presenting the suggestion in an interface associated with the operator graph.

13. The computer-implemented method of claim 12, further comprising, prior to joining the instruction code of the first processing element with the instruction code of the second processing element, receiving, via the interface, a selection of the first processing element and the second processing element in response to the suggestion;

wherein the at least one optimization rule specifies that the instruction code of the first processing element should be joined with the instruction code of the second processing element based on, in respective instances, each of (i) the type of the first processing element being compatible with the type of the second processing element and (ii) a function performed by the instruction code of the first processing element to a first field of a data tuple being compatible with a function performed by the instruction code of the second processing element to a second field of the data tuple.

14. The computer-implemented method of claim 13, wherein analyzing the plurality of processing elements comprises analyzing the instruction code of the first processing element relative to the instruction code of each of the other plurality of processing elements in the operator graph as the second processing element based on the at least one optimization rule;

wherein joining the first processing element with the second processing element comprises merging the instruction code of the first processing element and the instruction code of the second processing element into the third processing element;

wherein the computer-implemented method further comprising, subsequent to merging the instruction code of the first processing element and the second processing element into the third processing element, removing the first processing element and the second processing element from the operator graph, wherein the distributed application is deployed to the streams processing environment.

15. The computer-implemented method of claim 14, wherein the instruction code of the plurality of processing elements is generated by a compiler, wherein the instruction code of the plurality of processing elements comprises executable instruction code;

wherein generating the third processing element in effect at least partially reduces re-execution of functionally redundant code common to the first processing element and the second processing element.

16. The computer-implemented method of claim 1, further comprising, prior to joining the instruction code of the first processing element with the instruction code of the second processing element:

generating a suggestion to join the instruction code of the first processing element with the instruction code of the second processing element.

17. The computer-implemented method of claim 1, wherein the at least one optimization rule specifies that the instruction code of the first processing element should be joined with the instruction code of the second processing element based on the type of the first processing element being compatible with the type of the second processing element.

18. The computer-implemented method of claim 1, wherein the at least one optimization rule specifies that the instruction code of the first processing element should be joined with the instruction code of the second processing element based on a function performed by the instruction code of the first processing element being compatible with a function performed by the instruction code of the second processing element.

19. The computer-implemented method of claim 1, wherein the third processing element is generated by an integrated development environment (IDE) tool.

20. The computer-implemented method of claim 19, wherein the third processing element is generated by an optimization tool component of the IDE tool.