Linked Data is a design for sharing decentralized, but interrelated, data on the Web.
At any given point in the development lifecycle it is of great interest to understand how requirements, source code, test cases, and defects are related. Typical questions are,
- "Which requirements don't have any test cases?" and
- "How many defects are unresolved for each requirement?"
Design 1.
One design for integrating this lifecycle data is to develop relational
models for each type of development artifact and then store the
corresponding relational representations of all the artifacts in a data
warehouse where they can be queried, reported on, and analyzed using
conventional business intelligence tools such as BIRT and Cognos.
Analytical Reporting
Design 2.
Linked Data offers an alternative way to solve the data integration
problem. The key advance here is that Linked Data provides a uniform way
to identify artifacts, namely HTTP Uniform Resource Identifiers (URI),
and a common data model and representation format for them, Resource
Description Framework (RDF). Data integration among mutliple sources of development artifacts is
achieved by loading the RDF representations of all the development
artifacts into a shared triple store, e.g. Jena, which can be queried
using the powerful SPARQL query language.
The main tiers of this architecture are as follows:
- Data Source Tier - CLM Development Tools that provide Linked Data
- Reporting and Query Service Tier - Indexer, RDF Triple Store (Jena/TDB), SPARQL Endpoint
- Presentation Tier - Business Intelligence reporting and analysis (Cognos, BIRT), Document generation (RPE), Faceted browsing
Live Reporting
-
댓글 없음:
댓글 쓰기