Scheduler Web Service Invocation via Descriptors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing schedulers are limited to closed environments and cannot manage or invoke external services, particularly web services provided by third parties, which hinders the full exploitation of the internet and interoperability with legacy applications.
Innovation Solution
A scheduling method that associates work units with descriptors indicating external services, retrieves interface documents, builds request messages, and sends them to external services for invocation, enabling the scheduler to manage and integrate with external services in a distributed architecture.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a scheduler is designed for closed environments with single computer or cluster control, then it achieves reliable job management within the company network, but it cannot access or invoke external services provided by third parties
Solution Approach 1:
The patent introduces an intermediary layer (service descriptor and interface document) that bridges the scheduler and external services. The service descriptor acts as a mediator that contains metadata about external services, allowing the scheduler to interact with external services through a standardized interface while maintaining its原有 reliable job management capabilities within the closed environment.
Solution Approach 2:
The scheduler is enhanced with multi-functionality by adding the capability to both manage internal jobs (original function) and invoke external services (new function). This is achieved through a unified service descriptor mechanism that can represent both internal and external services, allowing the scheduler to operate universally across different service types without sacrificing its core reliability.
2Adaptability or versatility
If external services are accessed through various protocols and formats, then interoperability with web services is enabled, but the scheduler cannot effectively manage these services due to incompatible requirements
Solution Approach 1:
The patent applies parameter changes by transforming diverse external service protocols and formats into a unified service descriptor structure. The interface document standardizes the parameters and metadata of external services, allowing the scheduler to manage them through consistent operations regardless of the underlying protocol variations (HTTP, SOAP, XML, etc.).
Solution Approach 2:
The service management functionality is segmented into distinct components: service discovery, service descriptor creation, interface document retrieval, and service invocation. This segmentation allows each component to handle specific aspects of external service management independently, making the overall system easier to operate while maintaining broad interoperability.
3Manufacturing precision
If the scheduler invokes external services through complex interface documents and request messages, then precise service control is achieved, but the system complexity increases
Solution Approach 1:
The patent uses copying by creating service descriptors that contain metadata copies of external service interfaces. Instead of directly managing the complex original service interfaces, the scheduler works with simplified descriptor copies that capture the essential information needed for service invocation, thereby reducing system complexity while maintaining precise control.
Solution Approach 2:
The service descriptor and interface document are prepared in advance before actual service invocation. This preliminary action involves retrieving and storing service metadata, defining request message structures, and configuring service parameters beforehand. By performing these actions preliminarily, the actual service invocation process becomes simpler and more precise, reducing the perceived complexity during runtime.
Data Source
AI summary
The present invention proposes a method and a corresponding system for scheduling invocation of web services from a central point of control. A scheduler accesses a workload database, which associates an execution agent and a descriptor with each submitted job. The descriptor identifies a desired web service, an address of a corresponding WSDL document, and the actual content of a request message to be passed to the web service. Whenever the job is submitted for execution, the scheduler sends the job's descriptor to the associated agent. In response thereto, the agent downloads the WSDL document that specifies the structure of the messages supported by the web service. The scheduler builds a request message for the web service embedding the desired content into the structure specified in the WSDL document. The agent sends the request message to an endpoint implementing the web service, so as to cause its invocation.


