Systems and methods for debugging an application running on a parallel-processing computer system
A runtime system implemented in accordance with the present invention provides an application platform for parallel-processing computer systems. Such a runtime system enables users to leverage the computational power of parallel-processing computer systems to accelerate/optimize numeric and array-intensive computations in their application programs. This enables greatly increased performance of high-performance computing (HPC) applications.
1. A computer-implemented method, comprising:
at a computer having memory and a plurality of processing elements:
receiving one or more operation requests issued by an application and directed to a parallel-processing computer system, wherein the one or more operation requests include a first operation request and a second operation request consecutive to the first operation request, and the two consecutive operation requests are associated with a user-initiated data examination request;
preparing one or more compute kernels for the one or more operation requests, wherein the one or more compute kernels are configured to execute on the plurality of processing elements of the parallel-processing computer system, further including:
generating a first set of compute kernels corresponding to a merge of the first and second operation requests; and
generating a second set of compute kernels corresponding to the first operation request alone; and
arranging for execution of the one or more compute kernels on the plurality of processing elements, further including debugging the first set of compute kernels through the execution of the second set of compute kernels.
2. The method of claim 1 , wherein the user-initiated data examination request is embedded in the application.
3. The method of claim 1 , wherein the user-initiated data examination request is launched through a program debugging tool.
4. The method of claim 1 , wherein the user-initiated data examination request is a user-specified breakpoint.
5. The method of claim 1 , wherein results from the execution of the second set of compute kernels are used for debugging results from the execution of the first set of compute kernels.
6. A parallel-processing computer system, comprising:
memory;
a plurality of processing elements; and
at least one program stored in the memory and executed by the plurality of processing elements, the at least one program including:
instructions for receiving one or more operation requests issued by an application and directed to the parallel-processing computer system, wherein the one or more operation requests include a first operation request and a second operation request consecutive to the first operation request, and the two consecutive operation requests are associated with a user-initiated data examination request;
instructions for preparing one or more compute kernels for the one or more operation requests, wherein the one or more compute kernels are configured to execute on the plurality of processing elements of the parallel-processing computer system, further including:
instructions for generating a first set of compute kernels corresponding to a merge of the first and second operation requests; and
instructions for generating a second set of compute kernels corresponding to the first operation request alone; and
instructions for arranging for execution of the one or more compute kernels on the plurality of processing elements, further including instructions for debugging the first set of compute kernels through the execution of the second set of compute kernels.
7. The computer system of claim 6 , wherein the user-initiated data examination request is embedded in the application.
8. The computer system of claim 6 , wherein the user-initiated data examination request is launched through a program debugging tool.
9. The computer system of claim 6 , wherein the user-initiated data examination request is a user-specified breakpoint.
10. The computer system of claim 6 , wherein results from the execution of the second set of compute kernels are used for debugging results from the execution of the first set of compute kernels.
11. A computer program product for use in conjunction with a parallel-processing computer system, the computer program product comprising a computer readable storage medium and a computer program mechanism embedded therein, the computer program mechanism comprising:
instructions for receiving one or more operation requests issued by an application and directed to the parallel-processing computer system, wherein the one or more operation requests include a first operation request and a second operation request consecutive to the first operation request, and the two consecutive operation requests are associated with a user-initiated data examination request;
instructions for preparing one or more compute kernels for the one or more operation requests, wherein the one or more compute kernels are configured to execute on a plurality of processing elements of the parallel-processing computer system, further including:
instructions for generating a first set of compute kernels corresponding to a merge of the first and second operation requests; and
instructions for generating a second set of compute kernels corresponding to the first operation request alone; and
instructions for arranging for execution of the one or more compute kernels on the plurality of processing elements, further including instructions for debugging the first set of compute kernels through the execution of the second set of compute kernels.
12. The computer program product of claim 11 , wherein the user-initiated data examination request is embedded in the application.
13. The computer program product of claim 11 , wherein the user-initiated data examination request is launched through a program debugging tool.
14. The computer program product of claim 11 , wherein the user-initiated data examination request is a user-specified breakpoint.
15. The computer program product of claim 11 , wherein results from the execution of the second set of compute kernels are used for debugging results from the execution of the first set of compute kernels.
16. The computer program product of claim 11 , wherein the second set of compute kernels is not a subset of the first set of compute kernels.
17. The computer program product of claim 11 , wherein the second set of compute kernels is a subset of the first set of compute kernels.
18. The computer program product of claim 11 , wherein the first set of compute kernels is executed on one or more graphical processing units and the second set of compute kernels is executed on one or more central processing units.
19. The computer program product of claim 11 , wherein the first set of compute kernels includes at least one pre-existing compute kernel that was previously generated in response to one or more earlier operation requests issued by the application.
20. The method of claim 1 , wherein the second set of compute kernels is not a subset of the first set of compute kernels.
21. The method of claim 1 , wherein the second set of compute kernels is a subset of the first set of compute kernels.
22. The method of claim 1 , wherein the first set of compute kernels is executed on one or more graphical processing units and the second set of compute kernels is executed on one or more central processing units.
23. The method of claim 1 , wherein the first set of compute kernels includes at least one pre-existing compute kernel that was previously generated in response to one or more earlier operation requests issued by the application.
24. The computer system of claim 6 , wherein the second set of compute kernels is not a subset of the first set of compute kernels.
25. The computer system of claim 6 , wherein the second set of compute kernels is a subset of the first set of compute kernels.
26. The computer system of claim 6 , wherein the first set of compute kernels is executed on one or more graphical processing units and the second set of compute kernels is executed on one or more central processing units.
27. The computer system of claim 6 , wherein the first set of compute kernels includes at least one pre-existing compute kernel that was previously generated in response to one or more earlier operation requests issued by the application.