Self-Organizing Hive of Computing Engines

• US 20080263131A1
• Filed: 05/27/2008
• Published: 10/23/2008
• Est. Priority Date: 09/07/2002
• Status: Active Grant

• Current Assignee: Appistry Incorporated
• Original Assignee: Appistry Incorporated
• Inventors: Groner, Michael Scott, Hinni, Aaron Louis, Lozano, Roberto Raul, Semones, Guerry Anderson
• Granted Patent: US 8,060,552 B2
• Time in Patent Office: Days
• US Class Current: 709/202
• CPC Class Codes: H04L 41/00 Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
• CPC Class Codes: H04L 41/046 comprising network management agents or mobile agents therefor
• CPC Class Codes: H04L 67/01 Protocols
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46
• CPC Class Codes: H04L 67/51 Discovery or management thereof, e.g. service location protocol [SLP] or web services
• CPC Class Codes: H04L 69/16 Implementation or adaptation of Internet protocol [IP], of transmission control protocol [TCP] or of user datagram protocol [UDP]
• CPC Class Codes: H04L 69/163 In-band adaptation of TCP data exchange; In-band control procedures

Abstract

A self-organizing hive of computing engines is used to process information. A network communicatively couples multiple of hive engines together. These hive engines are configured to self-organize into a cooperative hive for performing computing of client jobs, with the cooperative hive including a multiple processing regions and a request region. This self-organization includes self-forming the processing regions and the request region by the hive engines such that each of the hive engines belongs to at least one of the processing regions; and wherein each the request region includes multiple of the plurality of hive engines.

Claims

1. A system for processing information comprising:

a network; and

a plurality of hive engines coupled to the network, wherein the plurality of hive engines are configured to self-organize into a cooperative hive for performing computing of client jobs, with the cooperative hive including a plurality of processing regions and a request region, wherein said self-organization includes self-forming the plurality of processing regions and the request region by the plurality of hive engines such that each of the plurality of hive engines belongs to at least one of the plurality of processing regions; and

wherein each the request region includes a plurality of the plurality of hive engines;

wherein each hive engine belonging to the request region includes;

means for responding to client requests from clients for processing client jobs;

means for assigning client jobs to hive engines of the plurality of hive engines, including means for requesting volunteer hive engines in at least one of the plurality of processing regions to process a client job and receiving volunteer responses;

means for receiving processing result information; and

means for communicating processing results to the clients based on said processing result information; and

wherein each hive engine belonging to one of the plurality of processing regions includes;

means for responding to volunteer requests for processing client jobs;

means for processing client jobs assigned to it; and

means for communicating said processing result information to the hive engine in the request region which assigned the corresponding client job to it.

2. The system of claim 1, wherein said means for processing client jobs by a hive engine belonging to one of the plurality of processing regions assigned to it includes:

means for sending partial processing state information for a particular client job to the hive engine in the request region which assigned the particular client job to it; and

wherein each said hive engine in the request region includes;

means for receive partial processing state information, and means for assigning continued of processing of a client job corresponding to and based on said received partial processing state information to another hive engine in one of the plurality of processing regions.

3. The system of claim 2, wherein said means for continued of processing of the client job is responsive to a detected failure with the processing of the client job by hive engine currently assigned to process the client job.

4. The system of claim 1, wherein each of hive engines belonging to one of the plurality of processing regions includes means for becoming a member of a processing region, including means for sending a multicast request to join a processing region message, and for selecting a processing region to join based on received responses to said multicast request to join a processing region message.

5. The system of claim 1, wherein each of the hive engines belonging to one of the plurality of processing regions includes means for forming a processing region, including means for sending a multicast request to join a processing region message;

and in response to receiving no responses, creating a new processing region.

6. The system of claim 1, wherein each of the plurality of hive engines belonging to a particular processing region of the plurality of processing regions includes means for selecting a processing region leader for leading self-organization tasks in the particular processing region from the hive engines belonging to the particular processing region.

7. The system of claim 6, wherein the processing region leader is configured to monitor and adjust the number of hive engines in the request region.

8. The system of claim 7, wherein said monitoring and adjusting the number of hive engines in the request region includes determining that there are too few hive engines in the request region, and in response, directing one of the hive engines in one of the plurality of processing regions to join the request region.

9. The system of claim 7, wherein said monitoring and adjusting the number of hive engines in the request region includes determining that there are too many hive engines in the request region, and in response, directing one of the hive engines in the request region to leave the request region.

10. The system of claim 6, wherein the processing region leader is configured to monitor and adjust the number of hive engines in one or more of the plurality of processing regions.

11. The system of claim 10, wherein said monitoring and adjusting the number of hive engines in one or more of the plurality of processing regions includes determining that there are too many hive engines in a particular processing region of the plurality of processing regions, and in response, causing the particular processing region to disband or split into multiple processing regions.

12. The system of claim 1, wherein the processing region leader is configured to send multicast heartbeat request messages to hive engines in the particular processing region and receiving responses from said hive engines in the particular processing region in order to determine an operational status of each of said hive engines in the particular processing region.

13. The system of claim 6, wherein each of the processing region leaders of the plurality of processing regions includes means for selecting a senior processing region leader for leading self-organization tasks in the plurality of processing regions.

14. The system of claim 13, wherein the senior processing region leader is configured to direct a selected one of said processing region leaders of the plurality of processing regions to leave the request region in response to determining the number of request handlers in the request region needs to be reduced.

15. The system of claim 13, wherein the senior processing region leader is configured to direct a selected one of said processing region leaders of the plurality of processing regions to disband or split a processing region of the plurality of processing regions in response to determining the number of processing regions needs to be adjusted.

16. The system of claim 1, wherein each of the plurality of hive engines is a computer.

17. The system of claim 1, wherein each hive engine includes:

means for starting up; and

means for joining a particular one of the plurality of processing regions.

18. The system of claim 1, wherein each hive engine belonging to the request region includes:

means for monitoring the progress of the client job; and

means for assigning continued processing of the client job from a partially completed state to a second hive engine belonging to one of the plurality of processing regions.

19. The system of claim 18, where the client job includes a reference or identification of a process flow, the process flow identifies one or more processing sets, each of said one or more processing sets identifying a beginning state, a task, and one or more ending states.

20. The system of claim 19, wherein each of the plurality of hive engines in at least one of the plurality of processing regions maintains a copy of the process flow.