Unbounded computing space

• US 20030225822A1
• Filed: 05/30/2002
• Published: 12/04/2003
• Est. Priority Date: 05/30/2002
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Zinda, Eric K., Olson, Lance E.
• Granted Patent: US 7,043,522 B2
• Time in Patent Office: Days
• US Class Current: 709/202
• CPC Class Codes: G06F 2209/505 Clust
• CPC Class Codes: G06F 9/5072 Grid computing

Abstract

An unbounded computing space including independent nodes is provided. Each node includes a discovery module to detect other nodes within the unbounded computing space, and an interaction module to communicate with the other detected nodes. Any node within the unbounded computing space can introduce an independent agent (e.g., an executable code assembly) into the unbounded computing space. The node in which an independent agent is introduced can subsequently send the independent agent to another node in the unbounded computing space for execution in the other node. Nodes in which an independent agent is introduced can define privileges that are associated with one or more agents and can be used to ensure the agent does not exceed a security policy as defined by the node. This allows for a distributed computing system in which an unlimited number of different agents can be introduced.

Claims

1. A distributed computing architecture, comprising:

a plurality of node modules in which agent programs can run, each node module comprising;

a discovery module that detects other node modules within the distributed computing architecture; and

an interaction module through which each node module communicates with other node modules,

wherein each node module is capable of introducing an agent program that can be sent to one or more other node modules, via the interaction module, and run in the one or more other node modules.

2. The distributed computing architecture of claim 1, wherein a discovery module of a first node communicates with a server that maintains a database of active node modules.

3. The distributed computing architecture of claim 1, wherein a discovery module of a first node broadcasts an undirected request over a network to detect other active node modules.

4. The distributed computing architecture of claim 1, wherein each node module executes in a managed code environment.

5. The distributed computing architecture of claim 4, wherein each node module executes the agent program based on a security policy predefined in each node.

6. The distributed computing architecture of claim 4, wherein the managed code environment comprises Common Language Runtime.

7. The distributed computing architecture of claim 4, wherein the managed code environment comprises a Java virtual machine.

8. The distributed computing architecture of claim 1, wherein each interaction module communicates with other node modules using HTTP messages.

9. The distributed computing architecture of claim 1, wherein each interaction module communicates with other node modules using TCP messages.

10. The distributed computing architecture of claim 1, wherein each node module defines virtual terrain, and each agent program defines a virtual terrain inhabitant.

11. The distributed computing architecture of claim 1, wherein the agent program coordinates a distributed task across the plurality of code modules.

12. The distributed computing architecture of claim 5, wherein each security module comprises a security policy of the managed code environment.

13. A method for executing a program on a computer system, comprising:

(a) introducing an agent program into a first node of a plurality of nodes, wherein each node provides an environment in which the agent program can run;

(b) detecting a second node of the plurality of nodes;

(c) sending the agent program to the second node; and

(d) executing the agent program in the second node.

14. The method of claim 13, wherein step (b) comprises:

(i) sending to a server a node list request; and

(ii) receiving from the server a list of active nodes.

15. The method of claim 13, wherein step (b) comprises broadcasting an undirected request over a network.

16. The method of claim 13, wherein step (c) comprises sending an HTTP message to the second node.

17. The method of claim 16, wherein in step (c) the HTTP message comprises POST data.

18. The method of claim 13, wherein step (c) comprises sending an XML message to the second node.

19. The method of claim 13, wherein step (d) comprises performing a diagnostic service on a local computer system.

20. The method of claim 19, wherein step (d) comprises:

(i) detecting an amount of available space in a storage device;

(ii) when the detected amount of available space is below a predetermined level, deleting unnecessary files meeting predefined criteria.

21. The method of claim 19, wherein step (d) comprises:

(i) detecting a level of file fragmentation in a storage device;

(ii) when the detected level of file fragmentation is above a predetermined level, initiating a defragmentation procedure on the storage device.

22. The method of claim 13, wherein further comprising the step of executing the agent program in the first node, wherein the agent program coordinates a distributed task across at least the first and second nodes.

23. In a computing architecture comprising a plurality of nodes that are capable of executing agent programs, and wherein each node comprises a detection module that detects other nodes in the computer architecture, and an interaction module that communicates with other nodes in the computer architecture, a method for distributing agent programs, comprising:

(a) introducing a first agent program by a first node of the plurality of nodes;

(b) according to a predetermined algorithm, determining a second node to which the first agent program should be sent;

(c) sending the first agent program to the second node;

(d) introducing a second agent program by a third node of the plurality of nodes;

(e) according to the predetermined algorithm, determining a fourth node to which the second agent program should be sent; and

(f) sending the second agent program to the fourth node.

24. The method of claim 23, wherein the predetermined algorithm comprises selecting a fourth node to which the agent program has not yet been sent.

25. The method of claim 23, wherein the predetermined algorithms comprises randomly selecting a fourth node.

26. A computer architecture comprising a first plurality of computing nodes, wherein any node within the first plurality of computing nodes is capable of introducing a program agent and sending the program agent to any other node within the first plurality of computing nodes, where the program agent subsequently runs.

27. The computer architecture of claim 26, wherein the computer architecture consists of the first plurality of nodes.

28. The computer architecture of claim 26, further comprising a second plurality of computing nodes, wherein any node within the second plurality of computing nodes is capable of introducing a second program agent and sending the second program agent to any other node within the second plurality of computing nodes, where the second program agent subsequently runs.

29. A method for performing a task across a distributed computer system, comprising:

(a) introducing an agent program into a first node of a plurality of nodes, wherein each of the plurality of nodes provides an environment in which the agent program can run and is a capable of receiving the agent program;

(b) dividing a task defined by the agent program into a plurality of sub-tasks;

(c) creating a plurality of derivative agent programs, wherein each derivative agent program is based on one sub-task of the plurality of sub-tasks; and

(d) sending each of the plurality of derivative agent programs to different nodes of the plurality of nodes.

30. The method of claim 29, wherein the selection of the different nodes is based on predetermined criteria.

31. The method of claim 30, wherein the predetermined criteria comprises a processor utilization level below a predetermined threshold.

32. The method of claim 30, wherein the predetermined criteria comprises a predetermined resource.