Granted Patent B2

US 9,158,584 · App. 13/856,102 · Granted Oct 13, 2015

Abstract

Claims (33)

1. A method for distributing sequential execution of a computer program to a plurality of hardware architectures of different types, the method comprising:

analyzing the computer program to identify a plurality of execution boundaries, wherein there is an execution boundary when the computer program transfers execution control to a library, and when execution control is transferred back to the computer program;

selecting one or more execution boundaries from the plurality of execution boundaries, based on predetermined parameters;

linking the computer program to the selected one or more execution boundaries;

executing the computer program with linked execution boundaries in an originating hardware architecture;

saving a hardware agnostic state of the execution of the computer program at a first execution stop point in the computer program, when the execution encounters a boundary from the selected one or more execution boundaries;

transmitting the hardware agnostic state to a remote hardware architecture to be executed on the remote hardware architecture, responsive to the hardware agnostic state; and

sequentially executing the computer program starting from the first execution stop point, by the remote hardware architecture, wherein the computer program is executed by a single hardware architecture at any point of time, during its execution.

2. The method of claim 1 , further comprising grouping the selected one or more boundaries for different types of hardware architectures; and linking the computer program to the grouping.

3. The method of claim 2 , further comprising saving a plurality of hardware agnostic states of the execution of the computer program responsive to the grouping; and transmitting the plurality of hardware agnostic states to a respective remote hardware architecture, responsive to the grouping.

4. The method of claim 1 , wherein analyzing the computer program further comprises determining which operating system and libraries are used by the computer program, determining what execution boundaries the identified operating system and libraries utilize within the computer program; and collecting boundary information for selecting one or more execution boundaries within the computer program.

5. The method of claim 4 , wherein the boundary information includes calling frequency of the execution boundary, and number of unique calls to the execution boundary.

6. The method of claim 5 , wherein the boundary information further includes call timing and calling dependency information.

7. The method of claim 1 , wherein the selected one or more execution boundaries are selected based on how, often they are used and how widespread their usage is.

8. The method of claim 1 , wherein the selected one or more execution boundaries are selected based on the type of the computer program.

9. The method of claim 1 , further comprising transferring the result of execution of the hardware agnostic state from the remote hardware to the originating hardware architecture.

10. The method of claim 1 , further comprising identifying a method of transmitting the hardware agnostic state to the remote hardware architecture.

11. The method of claim 1 , wherein the remote hardware architecture is selected randomly from a list of remote hardware architectures.

12. The method of claim 1 , wherein registers of the originating hardware architecture are converted to the registers of the hardware agnostic state and registers of the hardware agnostic state are converted to the registers of the remote hardware architecture.

13. A method for distributing execution of a computer program to a plurality of hardware architectures of different types, the method comprising:

analyzing the computer program to identify an execution boundary, wherein there is an execution boundary when the computer program transfers execution control to a library, and when execution control is transferred back to the computer program;

linking the computer program to the execution boundary;

executing the computer program with the linked execution boundary in an originating hardware architecture;

saving a hardware agnostic state of the execution of the computer program at a first execution stop point in the computer program, when the execution encounters said execution boundary;

transmitting the hardware agnostic state to a remote hardware architecture to be executed on the remote hardware architecture, responsive to the hardware agnostic state, wherein the remote hardware architecture is selected randomly from a list of remote hardware architectures; and

sequentially executing the computer program starting from the first execution stop point, by the remote hardware architecture, wherein the computer program is executed by a single hardware architecture at any point of time, during its execution.

14. The method of claim 13 , wherein the execution boundary includes calling frequency of the execution boundary, and number of unique calls to the execution boundary.

15. The method of claim 14 , wherein the execution boundary further includes call timing and calling dependency information.

16. The method of claim 13 , wherein the execution boundary is identified based on how often it is used and how widespread its usage is.

17. The method of claim 13 , wherein the execution boundaries is identified based on the type of the computer program.

18. The method of claim 13 , further comprising transferring the result of execution of the hardware agnostic state from the remote hardware to the originating hardware architecture.

19. The method of claim 13 , further comprising identifying a method of transmitting the hardware agnostic state to the remote hardware architecture.

20. The method of claim 13 , wherein registers of the originating hardware architecture are converted to the registers of the hardware agnostic state and registers of the hardware agnostic state are converted to the registers of the remote hardware architecture.

Assignments (15)