System and architecture for robust management of resources in a wide-area network

• US 8,572,290 B1
• Filed: 06/03/2011
• Issued: 10/29/2013
• Est. Priority Date: 05/02/2011
• Status: Active Grant

• Current Assignee: Board of Supervisors of Louisiana State University Agricultural & Mechanical College (Louisiana State University System)
• Original Assignee: Board of Supervisors of Louisiana State University Agricultural & Mechanical College (Louisiana State University System)
• Inventors: Mukhopadhyay, Supratik, Iyengar, S. Sitharama
• Primary Examiner(s): JAKOVAC, RYAN J
• Application Number:
  US13/153,388
• Time in Patent Office: 879 Days
• Field of Search: None
• US Class Current: 709/251
• CPC Class Codes: G06F 16/134 Distributed indices
• CPC Class Codes: G06F 16/27 Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: H04L 43/50 Testing arrangements
• CPC Class Codes: H04L 47/70 Admission control; Resource allocation
• CPC Class Codes: H04L 47/83 based on usage prediction

Abstract

A system and method of management of communication in a potentially unreliable wide-area network that contains one or more nodes connected to said network, each potentially having access to one or more inputs and/or outputs and capable of evaluating said inputs and directing said outputs, a global address space (GAS) accessible by said nodes, and a communication system using said GAS that provides communications between said nodes.

Claims

1. A system comprising:

a first networked node comprising;

a first computer, a first data store, a first control process, a first one or more logic processes, a first one or more evaluating processes, and a first communication system, said first control process having a capability to deploy said first one or more logic processes, and said first one or more logic processes each having a capability to cause said first one or more evaluating processes to begin,

said first data store, said first control process, said first one or more logic processes, said first evaluating process, and said first communication system operating on said first computer, and

a second networked node comprising;

a second computer, a second data store, a second control process, a second one or more logic processes, a second one or more evaluating processes, and a second communication system, said second control process having a capability to deploy said second one or more logic processes, and said second one or more logic processes each having a capability to cause said second one or more evaluating processes to begin,

said second data store, said second control process, said second one or more logic processes, said second one or more evaluating processes, and said second communication system operating on said second computer;

wherein said first data store provides access to a global address space and said second data store provides access to said global address space;

wherein data is available in said global address space;

wherein said first networked node and said second networked node are networked together;

wherein said first data store contains a first local copy of said data available in said global address space and said second data store contains a second local copy of said data available in said global address space;

wherein when first information of said data available in said global address space is altered in said first data store, said second local copy of said data available in said global address space in said second data store is updated correspondingly with altered said first information;

further comprising a plurality of additional networked nodes, each additional networked node comprising an additional computer and an additional data store, each of said plurality of additional data stores providing access to said global address space, wherein said first networked node, said second networked node, and said plurality of additional networked nodes are networked together;

wherein said plurality of additional data stores each contain an additional local copy of said data available in said global address space;

wherein when said first information of said data is altered in said first data store, each of said additional local copies is updated correspondingly with said altered first information;

wherein when second information of said data available in said global address space is altered in said second data store, said first local copy and said additional local copy in each of said plurality of additional data stores are updated correspondingly with said altered second information;

wherein said first data store, said second data store, and said plurality of additional data stores are configured to cause communications collectively to occur according to a pattern;

wherein said pattern is a ring;

wherein said first data store, said second data store, and said plurality of additional data stores each further comprises respectively a testing module, wherein said testing modules have the capability to collectively perform a test of resources available to a set of ringed computers selected from the group comprised of said first computer, said second computer, and said plurality of additional computers, each respective one of said set of computers associated with a respective one of said testing modules, and wherein an initiating one of said testing modules in said set of ringed computers has the capability to initiate said test for testing in other ringed computers in said set of ringed computers and to calculate a result based on said test; and

wherein each of said first data store, said second data store, and said plurality of additional data stores has a capability to temporarily prevent said communications from occurring.

3. The system according to claim 1, wherein each of said first data store, said second data store, and said plurality of additional data stores has a capability to use said result to determine whether to invoke said capability to temporarily prevent communications from occurring.

5. The system according to claim 3, wherein said test comprises the steps of:

(1) initiating said test and recording a current time in said initiating one of said testing modules;

(2) attempting transmission of one or more initiated test packets to a first one of said testing modules associated with a first other ringed computer;

(3) repeating, for each additional one of said testing modules of said other ringed computers;

(a) receiving a received one or more of said initiated test packets by a current one of said testing modules of said other ringed computers;

(b) measuring, for each received test packet, availability of resources to a current one of said other ringed computers associated with said current testing module;

(c) recording, in each of said received test packets, a resource measurement representing said availability of resources; and

(d) attempting transmission of each received test packet to a next one of said testing modules of said other ringed computers according to said pattern;

(4) receiving, by said initiating one of said testing modules, a set of returned test packets of said one or more initiated test packets;

(5) measuring, for each of said returned test packets, an amount of time representing time elapsed after initiation of said test; and

(6) calculating said result using said amount of time associated with each of said returned test packets and using each of said resource measurements associated with each of said returned test packets.

7. The system according to claim 3, wherein said first data store, said second data store, and said plurality of additional data stores each further comprising a learning module, wherein each of said learning modules has a capability to learn from said tests.

9. The system according to claim 5, wherein said first data store, said second data store, and said plurality of additional data stores each further comprising a learning module, wherein each of said learning modules has a capability to learn from said tests.

11. The system according to claim 7, wherein each of said learning modules has a capability to make predictions about resource utilization.

13. The system according to claim 9, wherein each of said learning modules has a capability to make predictions about resource utilization.

15. The system according to claim 11, wherein said first data store, said second data store, and said plurality of additional data stores have a capability to use said predictions in determining whether to temporarily prevent communications from being initiated.

17. The system according to claim 13, wherein said first data store, said second data store, and said plurality of additional data stores have a capability to use said predictions in determining whether to temporarily prevent communications from being initiated.

19. The system according to claim 7, wherein said learning module learns using an algorithm suitable for a high-dimensional setting.

21. The system according to claim 9, wherein said learning module learns using an algorithm suitable for a high-dimensional setting.

23. The system according to claim 15, wherein said learning module learns using an algorithm suitable for a high-dimensional setting.

25. The system according to claim 17, wherein said learning module learns using an algorithm suitable for a high-dimensional setting.

27. The system according to claim 19, wherein said algorithm is Vapnik'"'"'s Support Vector Machine algorithm.

29. The system according to claim 21, wherein said algorithm is Vapnik'"'"'s Support Vector Machine algorithm.

31. The system according to claim 23, wherein said algorithm is Vapnik'"'"'s Support Vector Machine algorithm.

33. The system according to claim 25, wherein said algorithm is Vapnik'"'"'s Support Vector Machine algorithm.

2. A system comprising:

a first networked node comprising;

a first computer, a first data store, a first control process, a first one or more logic processes, a first one or more evaluating processes, and a first communication system, said first control process having a capability to deploy said first one or more logic processes, and said first one or more logic processes each having a capability to cause said first one or more evaluating processes to begin,

said first data store, said first control process, said first one or more logic processes, said first evaluating process, and said first communication system operating on said first computer, and

a second networked node comprising;

a second computer, a second data store, a second control process, a second one or more logic processes, a second one or more evaluating processes, and a second communication system, said second control process having a capability to deploy said second one or more logic processes, and said second one or more logic processes each having a capability to cause said second one or more evaluating processes to begin,

said second data store, said second control process, said second one or more logic processes, said second one or more evaluating processes, and said second communication system operating on said second computer;

wherein said first data store provides access to a global address space and said second data store provides access to said global address space;

wherein data is available in said global address space;

wherein said first networked node and said second networked node are networked together;

wherein said first data store contains a first local copy of said data available in said global address space and said second data store contains a second local copy of said data available in said global address space;

wherein when first information of said data available in said global address space is altered in said first data store, said second local copy of said data available in said global address space in said second data store is updated correspondingly with altered said first information;

further comprising a plurality of additional networked nodes, each additional networked node comprising an additional computer and an additional data store, each of said plurality of additional data stores providing access to said global address space, wherein said first networked node, said second networked node, and said plurality of additional networked nodes are networked together;

wherein said plurality of additional data stores each contain an additional local copy of said data available in said global address space;

wherein when said first information of said data is altered in said first data store, each of said additional local copies is updated correspondingly with said altered first information;

wherein when second information of said data available in said global address space is altered in said second data store, said first local copy and said additional local copy in each of said plurality of additional data stores are updated correspondingly with said altered second information;

wherein said first data store, said second data store, and said plurality of additional data stores are configured to cause communications collectively to occur according to a pattern;

wherein said pattern is a multiring;

wherein said first data store, said second data store, and said plurality of additional data stores each further comprises respectively a testing module, wherein said testing modules have the capability to collectively perform a test of resources available to a set of ringed computers selected from the group comprised of said first computer, said second computer, and said plurality of additional computers, each respective one of said set of computers associated with a respective one of said testing modules, and wherein an initiating one of said testing modules in said set of ringed computers has the capability to initiate said test for testing in other ringed computers in said set of ringed computers and to calculate a result based on said test; and

wherein each of said first data store, said second data store, and said plurality of additional data stores has a capability to temporarily prevent said communications from occurring.

4. The system according to claim 2, wherein each of said first data store, said second data store, and said plurality of additional data stores has a capability to use said result to determine whether to invoke said capability to temporarily prevent communications from occurring.

6. The system according to claim 4, wherein said test comprises the steps of:

(1) initiating said test and recording a current time in said initiating one of said testing modules;

(2) attempting transmission of one or more initiated test packets to a first one of said testing modules associated with a first other ringed computer;

(3) repeating, for each additional one of said testing modules of said other ringed computers;

(a) receiving a received one or more of said initiated test packets by a current one of said testing modules of said other ringed computers;

(b) measuring, for each received test packet, availability of resources to a current one of said other ringed computers associated with said current testing module;

(c) recording, in each of said received test packets, a resource measurement representing said availability of resources; and

(d) attempting transmission of each received test packet to a next one of said testing modules of said other ringed computers according to said pattern;

(4) receiving, by said initiating one of said testing modules, a set of returned test packets of said one or more initiated test packets;

(5) measuring, for each of said returned test packets, an amount of time representing time elapsed after initiation of said test; and

(6) calculating said result using said amount of time associated with each of said returned test packets and using each of said resource measurements associated with each of said returned test packets.

8. The system according to claim 4, wherein said first data store, said second data store, and said plurality of additional data stores each further comprising a learning module, wherein each of said learning modules has a capability to learn from said tests.

10. The system according to claim 6, wherein said first data store, said second data store, and said plurality of additional data stores each further comprising a learning module, wherein each of said learning modules has a capability to learn from said tests.

12. The system according to claim 8, wherein each of said learning modules has a capability to make predictions about resource utilization.

14. The system according to claim 10, wherein each of said learning modules has a capability to make predictions about resource utilization.

16. The system according to claim 12, wherein said first data store, said second data store, and said plurality of additional data stores have a capability to use said predictions in determining whether to temporarily prevent communications from being initiated.

18. The system according to claim 14, wherein said first data store, said second data store, and said plurality of additional data stores have a capability to use said predictions in determining whether to temporarily prevent communications from being initiated.

20. The system according to claim 8, wherein said learning module learns using an algorithm suitable for a high-dimensional setting.

22. The system according to claim 10, wherein said learning module learns using an algorithm suitable for a high-dimensional setting.

24. The system according to claim 16, wherein said learning module learns using an algorithm suitable for a high-dimensional setting.

26. The system according to claim 18, wherein said learning module learns using an algorithm suitable for a high-dimensional setting.

28. The system according to claim 20, wherein said algorithm is Vapnik'"'"'s Support Vector Machine algorithm.

30. The system according to claim 22, wherein said algorithm is Vapnik'"'"'s Support Vector Machine algorithm.

32. The system according to claim 24, wherein said algorithm is Vapnik'"'"'s Support Vector Machine algorithm.

34. The system according to claim 26, wherein said algorithm is Vapnik'"'"'s Support Vector Machine algorithm.

35. A system comprising:

a first networked node comprising;

a first computer, a first data store, a first control process, a first one or more logic processes, a first one or more evaluating processes, and a first communication system, said first control process having a capability to deploy said first one or more logic processes, and said first one or more logic processes each having a capability to cause said first one or more evaluating processes to begin,

said first data store, said first control process, said first one or more logic processes, said first evaluating process, and said first communication system operating on said first computer, and

a second networked node comprising;

a second computer, a second data store, a second control process, a second one or more logic processes, a second one or more evaluating processes, and a second communication system, said second control process having a capability to deploy said second one or more logic processes, and said second one or more logic processes each having a capability to cause said second one or more evaluating processes to begin,

said second data store, said second control process, said second one or more logic processes, said second one or more evaluating processes, and said second communication system operating on said second computer;

wherein said first data store provides access to a global address space and said second data store provides access to said global address space;

wherein data is available in said global address space;

wherein said first networked node and said second networked node are networked together;

wherein said first communication system and said second communication system communicate via said global address space;

wherein said first one or more logic processes are capable of retrieving said data from said global address space via said first communication system;

wherein said second one or more logic processes are capable of retrieving said data from said global address space via said second communication system;

wherein said first control process is capable of deploying a first one of said first one or more logic processes with a first set of instructions;

wherein said first one of said first one or more logic processes is capable of causing a first one of said first one or more evaluating processes to begin executing said first set of instructions;

wherein said first one of said first one or more evaluating processes is capable of evaluating said data to synthesize a first result using said first set of instructions;

wherein said first one of said first one or more evaluating processes is capable of storing said first result as said data in said global address space via said first communication system;

wherein said first one of said first one or more logic processes is capable of causing a second one of said first one or more evaluating processes to begin executing said first set of instructions;

wherein said second one of said first one or more evaluating processes is capable of evaluating said data to synthesize a next result using said first set of instructions;

wherein said second one of said first one or more evaluating processes is capable of storing said next result as said data in said global address space via said first communication system; and

wherein said first one of said first one or more evaluating processes and said second one of said first one or more evaluating processes are executed concurrently on said first computer.