Selection of web services by service providers

• US 7,707,173 B2
• Filed: 07/15/2005
• Issued: 04/27/2010
• Est. Priority Date: 07/15/2005
• Status: Active Grant

• Current Assignee: X Corp. (f/k/a Twitter, Inc.) (X Holdings Corp.)
• Original Assignee: International Business Machines Corporation
• Inventors: Jaiswal, Suraj Kumar, Nanavati, Amit Anil, Chakraborty, Dipanjan, Misra, Archan
• Primary Examiner(s): Lu; Kuen S
• Assistant Examiner(s): Mitiku; Berhanu
• Application Number:
  US11/183,058
• Publication Number:
  US 20070016573A1
• Time in Patent Office: 1,747 Days
• Field of Search: 707/5, 707/100, 707/713, 707/790, 707/791, 705/26
• US Class Current: 707/608
• CPC Class Codes: G06F 16/958 Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 10/0633 Workflow analysis
• CPC Class Codes: H04L 67/51 Discovery or management thereof, e.g. service location protocol [SLP] or web services

Abstract

A Web services-oriented service provider middleware architecture and implementation is provided that integrates the performance monitoring of individual CPs, along with other dynamic contextual conditions, in the automatic selection of appropriate CPs. In the architecture, a CP'"'"'s performance is evaluated not only in terms of network or service-level parameters, but also via other business processes (e.g., complaint handling). The selection of multiple Web services occurs through an initial filtering of a set of feasible workflows for each task depending on business agreements and end user needs, and a subsequent dynamic context-based selection of the most appropriate workflow. A workflow filtering and ranking engine uses metarules to guide the optimisation process and help in selecting and ranking the feasible workflows. The component services in the workflows are evaluated by evaluator processes of information sources. The workflow engine outputs the workflows in their preference ranking. At runtime, if the workflow and/or its components meet all the runtime requirements (performance, for example), then the workflow is selected and executed.

Claims

1. A method for selecting a requested Web service from one of a plurality of providers, comprising:

defining a set of metarules, wherein each of said metarules comprises at least one condition and at least one action to be used in response to an instance of said at least one condition, said at least one condition comprising at least one parameter representing a context of a user and said at least one action comprising at least one directive for selecting services;

storing a set of alternative workflows for requested services, wherein each of said alternative workflows comprises multiple web services executed to accomplish a task and wherein each of said alternative workflows is associated with a different provider;

retaining performance information from previous web service invocations of said providers;

for each of said metarules, ranking said alternative workflows in said set of alternative workflows, based on said performance information, in order to output a set of context-dependent alternative workflows, including summaries of conditions under which each of said alternative workflows in said set of context-dependent alternative workflows should be executed;

receiving a customer service request from a customer;

matching context information from said customer with a summary of conditions for a first ranked one of said context-dependent alternative workflows; and

automatically selecting said first ranked one of said context-dependent alternative workflows for execution in response to said customer service request.

2. The method according to claim 1, wherein said performance information comprises a component based on technical service properties, and a component based on derived business properties.

3. The method according to claim 2, wherein said technical service properties include one or more of response time, availability, or latency.

4. The method according to claim 2, wherein said business properties include one or both of rankings and ratings of said providers.

5. The method according to claim 1, wherein said performance information comprises a long-term component that considers multiple instances of previous web service invocations, and an instance level component that considers individual service invocations.

6. The method according to claim 1, wherein said ranking comprises filtering said performance information according to a first metarule, parsing said set of feasible workflows according to said first metarule, and ranking parsed workflows according to filtered performance information.

7. The method according to claim 1, further comprising automatically selecting a next ranked one of said context-dependent alternative workflows for execution, if an immediately preceding ranked workflow fails to execute.

8. A method for selecting a requested Web service from one of a plurality of providers, comprising:

defining a set of metarules, wherein each of said metarules comprises at least one condition and at least one action to be used in response to an instance of said at least one condition, said at least one condition comprising at least one parameter representing a context of a user and said at least one action comprising at least one directive for selecting services;

storing a set of alternative workflows for requested services, wherein each of said alternative workflows comprises multiple web services executed to accomplish a task and wherein each of said alternative workflows is associated with a different provider;

retaining performance information from previous web service invocations;

for each of said metarules, ranking said alternative workflows in said set of alternative workflows, based on said performance information, in order to output a set of context-dependent alternative workflows, including summaries of conditions under which each of said alternative workflows in said set of context-dependent alternative workflows should be executed;

receiving a customer service request from a customer;

collecting run-time context data from said providers and said customer;

matching said run-time context data with a summary of conditions for a first ranked one of said context-dependent alternative workflows; and

automatically selecting said first ranked one of said context-dependent for execution in response to said customer service request.

9. The method according to claim 8, wherein said performance information comprises a component based on technical service properties, and a component based on derived business properties.

10. The method according to claim 9, wherein said technical service properties include one or more of response time, availability, or latency.

11. The method according to claim 9, wherein said business properties include one or both of rankings and ratings of said providers.

12. The method according to claim 8, wherein said performance information comprises a long-term component that considers multiple instances of previous web service invocations, and an instance level component that considers individual service invocations.

13. The method according to claim 9, wherein said ranking comprises filtering said performance information according to a first metarule, parsing said set of feasible workflows according to said first metarule, and ranking parsed workflows according to filtered performance information.

14. A web services selection server receiving service requests from customers and selecting a workflow from one of a plurality of providers, said web services selection server comprising:

a task database storing a set of alternative workflows for requested services, wherein each of said alternative workflows comprises multiple web services executed to accomplish a task and wherein each of said alternative workflows is associated with a different provider;

an evaluator module storing performance information from previous web service invocations by said providers;

a policies module storing metarules, wherein each of said metarules comprises at least one condition and at least one action to be used in response to an instance of said at least one condition, said at least one condition comprising at least one parameter representing a context of a user and said at least one action comprising at least one directive for selecting services;

a workflow engine ranking, for each of said metarules and based on said performance information, said alternative workflows in said set of alternative workflows in order to output a set of context-dependent alternative workflows, including summaries of conditions under which each of said alternative workflows in said set of context-dependent alternative workflows should be executed; and

an executor module receiving a customer service request from a customer, matching context information from said customer with a summary of conditions for a first ranked one of said context-dependent alternative workflows, and automatically selecting said first ranked one of said context-dependent alternative workflows for execution in response to said customer service request.

15. The web services selection server according to claim 14, said executor module further automatically selects a next ranked one of said context-dependent alternative workflows for execution, if an immediately preceding ranked workflow fails to execute.

16. A web services selection server receiving service requests from customers and selecting a workflow from one of a plurality of providers, said web services selection server comprising:

an offline module comprising;

a task database storing a set of alternative workflows for requested services, wherein each of said alternative workflows comprises multiple web services executed to accomplish a task and wherein each of said alternative workflows is associated with a different provider;

an evaluator module storing performance information from previous web service invocations by said providers;

a policies module storing metarules, wherein each of said metarules comprises at least one condition and at least one action to be used in response to an instance of said at least one condition, said at least one condition comprising at least one parameter representing a context of a user and said at least one action comprising at least one directive for selecting services;

a workflow engine ranking, for each of said metarules and based on said performance information, said alternative workflows in said set of alternative workflows in order to output a set of context-dependent alternative workflows, including summaries of conditions under which each of said alternative workflows in said set of context-dependent alternative workflows should be executed; and

a runtime module comprising;

a context server collecting run-time context data from said providers and said customers; and

an executor module receiving a customer service request from a customer, matching said run-time context data with a summary of conditions for a first ranked one of said context-dependent alternative workflows, and automatically selecting said first ranked one of said context-dependent alternative workflows for execution in response to said customer service request.