Systems and methods for parallel distributed programming

• US 8,473,906 B2
• Filed: 04/01/2010
• Issued: 06/25/2013
• Est. Priority Date: 05/21/2003
• Status: Active Grant

• Current Assignee: Regents of the University of California (University of California)
• Original Assignee: Regents of the University of California (University of California)
• Inventors: Pan, Lei, Bic, Lubomir R., Dillencourt, Michael B.
• Primary Examiner(s): Dam, Tuan Q
• Assistant Examiner(s): Wei, Zheng
• Application Number:
  US12/752,919
• Publication Number:
  US 20100257337A1
• Time in Patent Office: 1,181 Days
• Field of Search: None
• US Class Current: 717/119
• CPC Class Codes: G06F 8/314 Parallel programming languages G06F8/313 takes precedence
• CPC Class Codes: G06F 9/4862 the task being a mobile agent, i.e. specifically designed to migrate

Abstract

The present invention relates generally to computer programming, and more particularly to, systems and methods for parallel distributed programming. Generally, a parallel distributed program is configured to operate across multiple processors and multiple memories. In one aspect of the invention, a parallel distributed program includes a distributed shared variable located across the multiple memories and distributed programs capable of operating across multiple processors.

Claims

1. A computing system, comprising:

one or more memory areas configured to store one or more values corresponding to one or more logically shared variables; and

,

one or more processors configured to operate with one or more mobile agents of a computing program, the one or more mobile agents configured to migrate among the one or more processors configured to access the one or more values stored in the one or more memory areas, wherein the computing program includes a configuration to transform into a parallel computing program by forming one or more child computing programs to perform parallel operations, wherein the transform occurs responsive to a determination that at least one intermediate result will be used in execution of at least one of the child computing programs to be formed.

2. The system of claim 1, wherein the logically shared variables are distributed shared variables.

3. The system of claim 1, wherein the computing program and one or more child computing programs are distributed sequential computing programs.

4. The system of claim 1, wherein the parallel computing program is a distributed computing program.

5. The system of claim 1, wherein the one or more mobile agents are configured to be implemented with one or more self-migrating threads.

6. The system of claim 5, wherein the one or more self-migrating threads are configured to move from one processor to another processor.

7. The system of claim 5, wherein the one or more mobile agents are configured to operate within a shared memory system.

8. The system of claim 7, wherein the shared memory system is a distributed shared memory system.

9. The system of claim 1, wherein the one or more mobile agents are explicit-navigation mobile agents.

10. The system of claim 1, wherein the one or more mobile agents are implicit-navigation mobile agents.

11. The system of claim 1, wherein the one or more processors are located across a physical network.

12. The system of claim 1, wherein the distributed sequential computing program is configured to access the one or more logically shared variables.

13. The system of claim 1, wherein the parallel computing program is configured to operate across a distributed shared memory system.

14. The system of claim 1, wherein each of the one or more logically shared variables is a single logical variable that includes several variables that may be physically loaded into the one or more memory areas.

15. A distributed shared memory based system, comprising:

a physical network;

a logically shared variables memory system configured to access the physical network;

a mobile agent system coupled to the logically shared variables memory system; and

one or more processors located across the physical network and configured to operate with a parallel computing program configured to operate across the logically shared variables memory system, the parallel computing program comprising a sequential computing program and at least one child sequential computing programs spawned from the distributed sequential computing program and used concurrently to perform parallel processing and/or operations.

16. The system of claim 15, wherein the mobile agent system includes a configuration that supports compile-time explicit strong mobility.

17. The system of claim 15, wherein the mobile agent system includes a configuration that supports compile-time implicit strong mobility.

18. The system of claim 15, wherein the logically shared variables memory system is a distributed shared memory system.

19. The system of claim 15, wherein the sequential computing program and the at least one sequential computing program are distributed sequential computing programs comprising at least one mobile agent of the mobile agent system.

20. The system of claim 19, wherein the at least one mobile agent is configured to be implemented with self-migrating threads.

21. The system of claim 19, wherein the at least one mobile agent is an explicit-navigation mobile agent.

22. The system of claim 19, wherein the at least one mobile agent is an implicit-navigation mobile agent.

23. The system of claim 19, wherein the at least one mobile agent is configured to move from one processor to another processor.

24. The system of claim 19, wherein the at least one mobile agent operates within the logically shared variables memory system and wherein the logically shared variables memory system is a distributed shared memory system.

25. A computing system, comprising:

one or more memory means for storing one or more values corresponding to one or more logically shared variables; and

,

one or more processors means for operating with one or more mobile agents of a computing program, the one or more mobile agents configured to migrate among the one or more processors configured to access the one or more values stored in the one or more memory areas, wherein the computing program includes a configuration to transform into a parallel computing program by forming one or more child computing programs to perform parallel operations, wherein the transform occurs responsive to a determination that at least one intermediate result will be used in execution of at least one of the child computing programs to be formed.