COMPOSITION MODEL FOR CLOUD-HOSTED SERVING APPLICATIONS

• US 20110209007A1
• Filed: 02/23/2010
• Published: 08/25/2011
• Est. Priority Date: 02/23/2010
• Status: Active Grant

• Current Assignee: R2 Solutions LLC (Acacia Research Corporation)
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Makeev, Evgeniy, Banachowski, Scott Alexander, Chandra, Rohit, Stata, Raymond Paul, Lim, Swee, Budzinski, Jeffrey, Feng, Andrew An
• Granted Patent: US 9,229,980 B2
• Time in Patent Office: Days
• US Class Current: 714/48
• CPC Class Codes: G06F 11/0709 in a distributed system consisting of a plurality of standalone computer nodes, e.g. clusters, client-server systems
• CPC Class Codes: G06F 11/0715 in a system implementing multitasking multitasking per se G06F9/46
• CPC Class Codes: G06F 11/0793 Remedial or corrective actions recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40
• CPC Class Codes: G06F 16/24542 Plan optimisation

Abstract

Methods and apparatus for executing an application are disclosed. In accordance with one embodiment, a request is received. One or more of a plurality of module types are instantiated such that a plurality of module objects are generated. A query plan linking the plurality of module objects is executed such that a response to the request is generated. The response is then returned.

Claims

1. A method, comprising:

generating a query plan linking a plurality of module objects by instantiating one or more of a plurality of module types, wherein the plurality of module types includes a request module type and a response module type;

receiving a request;

executing the query plan linking the plurality of module objects such that a response to the request is generated, wherein an instance of the request module type processes the request and wherein the response is formed by an instance of the response module type; and

returning the response.

2. The method as recited in claim 1, wherein the plurality of module types further include an asynchronous module type, wherein the asynchronous module type invokes one or more network services.

3. The method as recited in claim 1, wherein the plurality of module types further includes a branch module type or a multibranch module type, wherein the branch module type determines one of two or more child modules to invoke, and wherein the multibranch module type selects one or more branch modules of two or more branch modules to invoke.

4. The method as recited in claim 1, wherein the plurality of module types further includes a fork module type whose child nodes are to be executed in parallel.

5. The method as recited in claim 1, wherein the plurality of module types further includes a join module type that performs synchronization for modules executing simultaneously on independent execution paths.

6. The method as recited in claim 1, wherein the plurality of module types further includes a termination module type that is executed when an error occurs during execution of the query plan.

7. The method as recited in claim 6, wherein the response is an error response.

8. The method as recited in claim 1, wherein the request is a search query.

9. The method as recited in claim 8, wherein the response includes search results.

10. The method as recited in claim 1, wherein the request initiates selection of one of a plurality of advertisements, and wherein the response includes a selected one of the plurality of advertisements.

11. A method, comprising:

generating one or more query plans by instantiating one or more of a plurality of module types such that a plurality of module objects are generated for each of the query plans, wherein the plurality of module types include at least one of an asynchronous module type, a branch module type, a multibranch module type, a fork module type, a join module type, or a termination module type;

receiving a request;

identifying one of the query plans to process the request;

executing the identified query plan such that a response to the request is generated; and

returning the response.

12. The method as recited in claim 11, further comprising:

validating input and output dependencies of the plurality of module objects within the query plan based upon interface definitions for the plurality of module objects.

13. The method as recited in claim 11, wherein the request is a search query.

14. The method as recited in claim 13, wherein the response includes search results.

15. The method as recited in claim 11, wherein the request initiates selection of one of a plurality of advertisements, and wherein the response includes a selected one of the plurality of advertisements.

16. The method as recited in claim 11, wherein the query plan is associated with a service being requested, wherein identifying one of the query plans to process the request comprises:

identifying a service associated with the request; and

identifying one of the query plans providing the service.

17. The method as recited in claim 11, wherein the request is received via a web server and the response is returned via the web server.

18. The method as recited in claim 11, wherein the request is one of a plurality of requests processed in parallel, wherein each of the plurality of requests is processed within one of a plurality of threads.

19. A system, comprising:

means for instantiating one or more query plans such that each of the query plans links a corresponding plurality of module objects;

means for instantiating one or more of a plurality of module types such that a plurality of module objects are generated;

means for receiving a request;

means for executing at least one of the query plans linking a corresponding plurality of module objects such that a response to the request is generated; and

means for returning the response.

20. The system as recited in claim 19, wherein the plurality of module types comprise at least one of an asynchronous module type, a branch module type, a multibranch module type, a fork module type, or a join module type.

21. The system as recited in claim 20, wherein the asynchronous module type invokes an external service and wherein the asynchronous module type includes a callback method that is called by a horizontal container when a response is received from the external service by the horizontal container.

22. The system of claim 20, further comprising:

means for identifying one of the query plans to execute in order to generate a response to a request.

23. A system, comprising:

one or more processors; and

one or more memories, wherein the processors and the memories are configured to provide;

a plurality of module types;

middleware configured to generate one or more query plans by instantiating one or more of the plurality of module types such that a plurality of module objects are generated for each of the query plans, identify one of the query plans to generate a response to a request, and execute the identified query plan; and

a container configured to receive a request and forward the request to the middleware, and configured to receive a response from the middleware and return the response.

24. The system as recited in claim 23, wherein the container is further configured to communicate one or more endpoint definitions to the middleware, where the endpoint definitions are mapped to one or more endpoints in a cloud.

25. The system as recited in claim 24, wherein the system further comprises:

a cloud, wherein the cloud communicates one or more endpoints in the cloud to the container.