System and Method of Planning for Cooperative Information Processing

• US 20080256549A1
• Filed: 05/11/2007
• Published: 10/16/2008
• Est. Priority Date: 04/10/2007
• Status: Abandoned Application

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Riabov, Anton V., Liu, Zhen, Ye, Fan
• Application Number:
  US11/747,826
• Publication Number:
  US 20080256549A1
• Time in Patent Office: Days
• US Class Current: 718/106
• CPC Class Codes: G06F 2209/5017 Task decomposition
• CPC Class Codes: G06F 9/4881 Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
• CPC Class Codes: G06F 9/5027 the resource being a machine, e.g. CPUs, Servers, Terminals
• CPC Class Codes: G06F 9/5066 Algorithms for mapping a plurality of inter-dependent sub-tasks onto a plurality of physical CPUs mappping at compile time, see G06F8/451
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling

Abstract

A cooperative data stream processing system is provided that utilizes a plurality of independent, autonomous and possibly heterogeneous sites in a cooperative arrangement to execute jobs derived from user-define inquires over dynamic, continuous streams of data. A method is provided for cooperative planning for the execution of the jobs across the distributed plurality of sites. An identification of the resources available for sharing from each one of the plurality of sites is communicated to one or more planners disposed on the distributed sites. These planners use the resource information to generate planning domains in which the jobs can be processed. Upon receipt of an inquiry at one of the sites, the inquiry is communicated to one of the planners that uses the planning domain to create at least one distributed plan for the inquiry. Processing of the inquiry is conducted in accordance with the distributed plan. Planning can take advantage of the structure of virtual organizations including cooperative and federated virtual organizations. The distributed plans can make use of the resources within a single virtual organization or across multiple organizations.

Claims

1. A method for planning cooperative processing in a cooperative data stream processing system, the method comprising:

identifying a plurality of distributed sites, each site comprising components capable of independently processing continuous dynamic streams of data;

communicating an identification of resources disposed on each site and available for sharing among the plurality of distributed sites to at least one planner disposed on one of the identified plurality of sites;

communicating an inquiry to be processed over the continuous dynamic streams of data to the planner; and

creating a distributed plan comprising the identified resources for processing the inquiry.

2. The method of claim 1, wherein the resources available for sharing comprise computational resources, software components, data sources, hardware resources or combinations thereof.

3. The method of claim 1, further comprising establishing common interest policies among the plurality of distributed sites that identify resources available for inter-site sharing.

4. The method of claim 1, wherein:

the method further comprises receiving an inquiry at a first site within the plurality of identified sites forming a cooperative virtual organization; and

the planner to which the resources available for sharing are communicated is disposed on the first site.

5. The method of claim 1, wherein:

the method further comprises receiving an inquiry at a first site within the plurality of identified sites; and

the planner to which the resources available for sharing is communicated is disposed on a lead site within a federated virtual organization of sites containing the first site.

6. The method of claim 1, further comprising constructing a planning domain using the communicated identification of resources available for sharing, the constructed planning domain comprising an identification of resources available for sharing, inter-site communication networks configurations, type systems, ontologies, security policy information or combinations thereof.

7. The method of claim 6, wherein the step of creating the distributed plan further comprises using the constructed planning domain to create the distributed plan.

8. The method of claim 1, further comprising associating a cost with each resource for a use of each resource available for sharing by one of the sites within the plurality of identified distributed sites;

and

the step of creating the distributed plan further comprises using the associated costs.

9. The method of claim 1, wherein:

the step of creating the distributed plan further comprises creating a plurality of distributed plans; and

the method further comprises selecting one of the plurality of distributed plans to be used in processing the communicated inquiry.

10. The method of claim 9, wherein:

the method further comprises associating a plurality of computed metrics with each created distributed plan; and

the step of selecting one of the plurality of distributed plans further comprises using the associated metrics to identify one of the plurality of distributed plans to be selected.

11. The method of claim 10, wherein the metrics comprise utility, quality, cost, inter-site communication bandwidth requirements, central processing unit consumption requirements or combinations thereof.

12. The method of claim 9, wherein each one of the plurality of distributed plans comprises a plurality of sub-plans and each one of the plurality of sub-plans comprises a job capable of being executed at one site within a virtual organization.

13. The method of claim 9, further comprising using the planner to establish agreements to secure required resources from identified resources available for sharing.

14. The method of claim 9, wherein:

the site containing the planner to which the inquiry has been communicated is a member of a plurality of virtual organizations, each virtual organization comprising a combination of sites selected from the identified plurality of distributed sites and configured to share at least one of data and processing resources within the combination of sites; and

the step of creating the plurality of distributed plans further comprises creating a separate distributed plan for each virtual organization to which the site containing the planner is a member.

15. The method of claim 14, wherein a given distributed plan from the plurality of distributed plans contains resources only from the virtual organization associated with the given distributed plan.

16. The method of claim 14, wherein one or more of the plurality of distributed plans comprises resources from two or more of the virtual organizations to which the site containing the planner is a member.

17. The method of claim 12, further comprising:

deploying each one of the separate distributed plans on the virtual organization for which each plan was created;

communicating results from each one of the separate, deployed distributed plans to the site containing the planner; and

combining the communicated results at the site containing the planner to create a single combined set of results.

18. The method of claim 1, wherein the step of communicating an identification of resources further comprises communicating the identification of resources to each one of a plurality of planners disposed on a plurality of distinct sites selected from the plurality of identified distributed sites.

19. The method of claim 1, wherein:

the site on which the planner is disposed is a lead site within a federated virtual organization; and

the step of communicating an inquiry further comprises communicating each one of a plurality of inquiries to the lead site planner.

20. The method of claim 1, wherein the step of creating the distributed plan, further comprises creating a back-up distributed plan to be used when the distributed plan cannot be deployed and creating at least one contingency distributed plan comprising alternative plan fragments to replaced failed plan fragments within the distributed plan.