Interactive partitioning and mapping of an application across multiple heterogeneous computational devices from a co-simulation design environment

• US 9,442,696 B1
• Filed: 01/16/2014
• Issued: 09/13/2016
• Est. Priority Date: 01/16/2014
• Status: Active Grant

• Current Assignee: The MathWorks, Inc.
• Original Assignee: The MathWorks, Inc.
• Inventors: Koh, David, Belge, Murat
• Primary Examiner(s): Dao, Thuy
• Assistant Examiner(s): Rivera, Anibal
• Application Number:
  US14/157,239
• Time in Patent Office: 971 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 2117/08 HW-SW co-design, e.g. HW-SW partitioning
• CPC Class Codes: G06F 30/20 Design optimisation, verification or simulation optimisation, verification or simulation of circuit designs G06F30/30
• CPC Class Codes: G06F 30/331 with hardware acceleration, e.g. by using field programmable gate array [FPGA] or emulation
• CPC Class Codes: G06F 8/00 Arrangements for software engineering testing or debugging G06F11/36; administrative, planning or organisation aspects of software project management G06Q10/06
• CPC Class Codes: G06F 8/34 Graphical or visual programming
• CPC Class Codes: G06F 8/447 Target code generation
• CPC Class Codes: G06F 9/455 Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
• CPC Class Codes: G06F 9/5044 considering hardware capabilities
• CPC Class Codes: G06F 9/5066 Algorithms for mapping a plurality of inter-dependent sub-tasks onto a plurality of physical CPUs mappping at compile time, see G06F8/451

Abstract

In an embodiment, a method for interactively changing allocation of an application including multiple algorithm components executing on a heterogeneous target environment (HTE) provides a user interface in a co-simulation design environment. The user interface is associated with the application having multiple algorithm components executing on the HTE that includes multiple computing devices with different processing capabilities. The method also sets attributes of the allocation scheme of the application using the user interface. The setting occurs when the application is executing. The method further receives data associated with the executing of the application in the co-simulation design environment when the application is executing subsequent to the setting of the attributes of the allocation scheme.

Claims

1. One or more non-transitory computer-readable storage media storing:

one or more instructions that, when executed, cause at least one processor to;

interact with a co-simulation design environment, the interacting for communicating with a target environment, the target environment including a plurality of heterogeneous computational devices;

partition, during co-simulation of an application, the application into a plurality of components;

allocate, using the co-simulation design environment, execution of a selected component of the application from among the plurality of components to a selected heterogeneous computational device of the target environment;

wherein the plurality of components execute on the plurality of heterogeneous computational devices;

receive an interactive instruction via the co-simulation design environment, the interactive instruction to alter allocation of the execution of the selected component of the application; and

alter, based on the interactive instruction, the allocation of the execution of the selected component from the selected heterogeneous computational device to another one of the plurality of heterogeneous computational devices during co-simulation of the application.

2. The medium of claim 1, further storing:

one or more instructions that, when executed, cause the at least one processor to;

determine one or more input/output (I/O) interfaces for the plurality of heterogeneous computational devices, the plurality of heterogeneous computational devices communicate with each other via the I/O interfaces.

3. The medium of claim 2, further storing:

one or more instructions that, when executed, cause the at least one processor to;

alter the one or more I/O interfaces associated with at least one of the plurality of heterogeneous computational devices.

4. The medium of claim 1, wherein the allocating and altering are performed without performing one or more of:

re-generating, re-compiling, re-synthesizing or re-running code for the application.

5. The medium of claim 1, wherein the plurality of heterogeneous computational devices include two or more of a processor, a simulator of a processor, a hardware description language (HDL) simulator, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a graphical programming unit (GPU), a programmable logic controller (PLC), a SystemC simulator, a digital signal processor (DSP) and a transaction-level model (TLM) based virtual platform.

6. The medium of claim 1, wherein the altering is based on an input received from an input device or a program.

7. The medium of claim 1, further storing:

one or more instructions that, when executed, cause the at least one processor to;

receive execution data from the selected component executing on the selected heterogeneous computational device, the receiving occurring;

in the co-simulation design environment, and

while the application is executing.

8. The medium of claim 7, wherein the altering is based on the received execution data.

9. The medium of claim 7, further storing:

one or more instructions that, when executed, cause the at least one processor to;

display the received execution data in the co-simulation design environment during the executing of the selected component on the selected heterogeneous computational device of the target environment; and

receive subsequent data in the co-simulation design environment during the executing, the subsequent data associated with the executing of the application following the altering the allocation of the execution of the selected component.

10. The medium of claim 7, wherein the received data associated with the executing of the selected component includes statistical data relating to the executing of the application on the target environment.

11. The medium of claim 10, wherein the statistical data includes data relating to at least one of:

a buffer allocation, algorithm data synchronization information, a wait time between execution of the selected component, power consumption information, a gate count, a transaction count, a processing element count, and CPU utilization by the plurality of heterogeneous computational devices.

12. The medium of claim 1, further storing:

one or more instructions that, when executed, cause the at least one processor to;

generate a fixed mapping between the selected component and the selected heterogeneous computational device.

13. The medium of claim 12, further storing:

one or more instructions that, when executed, cause the at least one processor to;

determine, based on a pre-determined criteria, that the fixed mapping is an optimal fixed mapping.

14. The medium of claim 1, further storing:

one or more instructions that, when executed, cause the at least one processor to;

generate a conventional dynamic allocator.

15. The medium of claim 14, wherein the conventional dynamic allocator implements a rule set, the rule set:

including one or more rules, and

used in allocating the execution of the selected component to the selected heterogeneous computational device.

16. The medium of claim 1, wherein the allocating is based on a plurality of rule sets, the plurality of rule sets being organized in a hierarchy.

17. The medium of claim 16, storing:

one or more instructions that, when executed, cause the at least one processor to;

generate an optimal conventional dynamic allocator implementing the plurality of rule sets.

18. The medium of claim 16, wherein:

the plurality of rule sets includes an optimal rule set, and

an optimal conventional dynamic allocator implements the optimal rule set.

19. The medium of claim 1, wherein the allocating maintains data synchronization within the application.

20. The medium of claim 1, wherein the application includes a graphical model, the graphical model including a plurality of blocks, at least one of the plurality of blocks corresponding to the selected component.

21. The medium of claim 1, wherein the application includes a text-based model, the text-based model including a plurality of functions or scripts, at least one of the plurality of functions or scripts corresponding to the selected component.

22. One or more non-transitory computer-readable storage media storing:

one or more instructions that, when executed, cause at least one processor to;

interact with a co-simulation design environment, the interacting to communicate with an application executing on a target environment, the target environment including a plurality of heterogeneous computational devices;

partition the application into one or more components during co-simulation of the application;

allocate executing of the one or more components to the plurality of heterogeneous computational devices;

receive execution data from at least one of the one or more components in the co-simulation design environment while the application is running; and

disable the executing of at least one of the one or more components of the application on one of the plurality of heterogeneous computational devices when the application is executing on the plurality of heterogeneous computational devices based on the received execution data.

23. The medium of claim 22, further storing:

one or more instructions that, when executed, cause the at least one processor to;

re-enable the executing of the at least one of the one or more components on the one of the plurality of the heterogeneous computational devices.

24. The medium of claim 23, wherein the re-enabling occurs when the application is executing on the plurality of heterogeneous computational devices, the re-enabling occurring without performing one or more of:

re-generating, re-compiling, re-synthesizing or re-running code for the application.

25. The medium of claim 22, further storing:

one or more instructions that, when executed, cause the at least one processor to;

receive execution data from the at least one of the one or more components executing on the one of the plurality of heterogeneous computational devices, the receiving occurring;

in the co-simulation design environment, and

when the application is executing.

26. The medium of claim 25, wherein the disabling is based on the received execution data.

27. The medium of claim 22, further storing:

one or more instructions that, when executed, cause the at least one processor to;

determine one or more input/output (I/O) interfaces for the plurality of heterogeneous computational devices, the plurality of heterogeneous computational devices communicate with each other via the I/O interfaces.

28. The medium of claim 27, further storing:

one or more instructions that, when executed, cause the at least one processor to;

alter the one or more I/O interfaces associated with at least one of the plurality of heterogeneous computational devices.

29. A computer-implemented method comprising:

interacting with a co-simulation design environment, the interacting to communicate with a target environment, the target environment including a plurality heterogeneous computational devices;

partitioning, during co-simulation of the application, the application into a plurality of components;

allocating execution of at least one of the plurality of components to one or more of the plurality of heterogeneous computational devices;

receiving execution data from at least one of the plurality of components in the co-simulation design environment while the application is running;

receiving an interactive instruction via the co-simulation design environment, the interactive instruction to alter allocation of the execution of the at least one the plurality of components based on the execution data;

altering, based on the received instruction, the allocation of the execution of the at least one the plurality of components from the one or more of the plurality of heterogeneous computational devices to a different one of the plurality of heterogeneous computational devices during co-simulation of the application.

30. A computer-implemented method comprising:

interacting with a co-simulation design environment, the interacting to communicate with an application executing on a target environment, the target environment including heterogeneous computational devices;

partitioning the application into one or more components during co-simulation of the application;

allocating, based on the partitioning, execution of the one or more components to one or more of the heterogeneous computational devices during the execution of the application; and

receiving execution data from at least one of the one or more components in the co-simulation design environment while the application is running; and

disabling the execution of at least one of the one or more components on at least one of the heterogeneous computational devices during execution of the application based on the received execution data.

31. A system for interactively changing allocation of an application executing in a target environment, comprising:

a processor interacting with the memory to;

interact with a co-simulation design environment, the co-simulation design environment performing co-simulation of an application executing on a target environment,

communicate with the application while the application executes on the target environment, the communicating based on the interacting, the target environment including heterogeneous computational devices;

partition the application into a plurality of components;

allocate execution of at least one of the plurality of components to one or more of the heterogeneous computational devices, where the allocating occurs when without performing one or more of;

re-generating, re-compiling, re-synthesizing or re-running code for the application;

receive an interactive instruction via the co-simulation design environment, altering allocation of the execution of the at least one of the plurality of components of the application while the application is executing, the altering based on the received interactive instruction;

receive execution data from the application executing on the one or more heterogeneous computational devices, the receiving occurring;

in the co-simulation design environment, and

when the application is executing; and

alter allocation of the execution of the at least one of the plurality of components of the application based on the received data, the altering performed while the application is executing on the heterogeneous computational devices, the allocating and altering occurring without performing one or more of;

re-generating, re-compiling, re-synthesizing or re-running code for the application.