IPC분류정보
국가/구분 |
United States(US) Patent
등록
|
국제특허분류(IPC7판) |
|
출원번호 |
US-0189336
(2008-08-11)
|
등록번호 |
US-8495603
(2013-07-23)
|
발명자
/ 주소 |
- Archer, Charles J.
- Blocksome, Michael A.
- Ratterman, Joseph D.
- Sidelnik, Albert
- Smith, Brian E.
|
출원인 / 주소 |
- International Business Machines Corporation
|
대리인 / 주소 |
|
인용정보 |
피인용 횟수 :
8 인용 특허 :
34 |
초록
▼
Methods, apparatus, and products are disclosed for generating an executable version of an application using a distributed compiler operating on a plurality of compute nodes that include: receiving, by each compute node, a portion of source code for an application; compiling, in parallel by each comp
Methods, apparatus, and products are disclosed for generating an executable version of an application using a distributed compiler operating on a plurality of compute nodes that include: receiving, by each compute node, a portion of source code for an application; compiling, in parallel by each compute node, the portion of the source code received by that compute node into a portion of object code for the application; performing, in parallel by each compute node, inter-procedural analysis on the portion of the object code of the application for that compute node, including sharing results of the inter-procedural analysis among the compute nodes; optimizing, in parallel by each compute node, the portion of the object code of the application for that compute node using the shared results of the inter-procedural analysis; and generating the executable version of the application in dependence upon the optimized portions of the object code of the application.
대표청구항
▼
1. A method of generating an executable version of an application using a distributed compiler operating on a plurality of compute nodes, the method comprising: receiving, by each compute node, a portion of source code for an application;compiling, in parallel by each compute node, the portion of th
1. A method of generating an executable version of an application using a distributed compiler operating on a plurality of compute nodes, the method comprising: receiving, by each compute node, a portion of source code for an application;compiling, in parallel by each compute node, the portion of the source code received by that compute node into a portion of object code for the application;performing, in parallel by each compute node, inter-procedural analysis on the portion of the object code of the application for that compute node, including sharing results of the inter-procedural analysis among the compute nodes, wherein sharing the results among the compute nodes further comprises performing by the compute nodes an All-To-All collective operation;optimizing, in parallel by each compute node, the portion of the object code of the application for that compute node using the shared results of the inter-procedural analysis including: identifying internal functions and external functions in the portion of the object code of the application for that compute node, wherein identifying external functions includes determining that a function in the portion of the object code is invoked by object code portions on another compute node using the shared results of the inter-procedural analysis; andoptimizing the portion of the object code of the application for that compute node in dependence upon the internal functions and the external functions for that portion of the object code; andgenerating the executable version of the application in dependence upon the optimized portions of the object code of the application. 2. The method of claim 1 wherein sharing results of the inter-procedural analysis among compute nodes further comprises sharing pointer aliasing information among the compute nodes. 3. The method of claim 1 wherein sharing results of the inter-procedural analysis among compute nodes further comprises sharing global variable usage information among the compute nodes. 4. The method of claim 1 wherein sharing results of the inter-procedural analysis among compute nodes further comprises sharing function argument value information among the compute nodes. 5. The method of claim 1 wherein generating an executable version of an application using a distributed compiler operating on a plurality of compute nodes further comprises generating the executable version of the application using the distributed compiler operating on the plurality of compute nodes of a parallel computer, the plurality of compute nodes connected for data communications using a plurality of data communications networks, at least one of the data communications networks optimized for point to point operations, and at least one of the data communications networks optimized for collective operations. 6. A parallel computer for generating an executable version of an application using a distributed compiler operating on a plurality of compute nodes, the parallel computer comprising one or more computer processors and computer memory operatively coupled to the computer processors, the computer memory having disposed within it computer program instructions configured to: receiving, by each compute node, a portion of source code for an application;compiling, in parallel by each compute node, the portion of the source code received by that compute node into a portion of object code for the application;performing, in parallel by each compute node, inter-procedural analysis on the portion of the object code of the application for that compute node, including sharing results of the inter-procedural analysis among the compute nodes, wherein sharing the results among the compute nodes further comprises performing by the compute nodes an All-To-All collective operation;optimizing, in parallel by each compute node, the portion of the object code of the application for that compute node using the shared results of the inter-procedural analysis including: identifying internal functions and external functions in the portion of the object code of the application for that compute node, wherein identifying external functions includes determining that a function in the portion of the object code is invoked by object code portions on another compute node using the shared results of the inter-procedural analysis; andoptimizing the portion of the object code of the application for that compute node in dependence upon the internal functions and the external functions for that portion of the object code; andgenerating the executable version of the application in dependence upon the optimized portions of the object code of the application. 7. The parallel computer of claim 6 wherein sharing results of the inter-procedural analysis among compute nodes further comprises sharing pointer aliasing information among the compute nodes. 8. The parallel computer of claim 6 wherein sharing results of the inter-procedural analysis among compute nodes further comprises sharing global variable usage information among the compute nodes. 9. The parallel computer of claim 6 wherein sharing results of the inter-procedural analysis among compute nodes further comprises sharing function argument value information among the compute nodes. 10. The parallel computer of claim 6 wherein the plurality of compute nodes are connected for data communications using a plurality of data communications networks, at least one of the data communications networks optimized for point to point operations, and at least one of the data communications networks optimized for collective operations. 11. A computer program product for generating an executable version of an application using a distributed compiler operating on a plurality of compute nodes, the computer program product comprising a non-transmission, computer readable medium, the non-transmission, computer readable medium comprising computer program instructions configured to: receiving, by each compute node, a portion of source code for an application;compiling, in parallel by each compute node, the portion of the source code received by that compute node into a portion of object code for the application;performing, in parallel by each compute node, inter-procedural analysis on the portion of the object code of the application for that compute node, including sharing results of the inter-procedural analysis among the compute nodes, wherein sharing the results among the compute nodes further comprises performing by the compute nodes an All-To-All collective operation;optimizing, in parallel by each compute node, the portion of the object code of the application for that compute node using the shared results of the inter-procedural analysis including: identifying internal functions and external functions in the portion of the object code of the application for that compute node, wherein identifying external functions includes determining that a function in the portion of the object code is invoked by object code portions on another compute node using the shared results of the inter-procedural analysis; andoptimizing the portion of the object code of the application for that compute node in dependence upon the internal functions and the external functions for that portion of the object code; andgenerating the executable version of the application in dependence upon the optimized portions of the object code of the application. 12. The computer program product of claim 11 wherein sharing results of the inter-procedural analysis among compute nodes further comprises sharing pointer aliasing information among the compute nodes. 13. The computer program product of claim 11 wherein sharing results of the inter-procedural analysis among compute nodes further comprises sharing global variable usage information among the compute nodes. 14. The computer program product of claim 11 wherein sharing results of the inter-procedural analysis among compute nodes further comprises sharing function argument value information among the compute nodes. 15. The computer program product of claim 11 wherein generating an executable version of an application using a distributed compiler operating on a plurality of compute nodes further comprises generating the executable version of the application using the distributed compiler operating on the plurality of compute nodes of a parallel computer, the plurality of compute nodes connected for data communications using a plurality of data communications networks, at least one of the data communications networks optimized for point to point operations, and at least one of the data communications networks optimized for collective operations.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.