TopBraid Platform Overview
TopBraid packages a stack of technologies that provide a complete solution platform for building services and applications that operate on Linked Data. The platform builds on RDF data model and SPARQL, the query language for RDF. It turns SPARQL into an extremely powerful tool for specifying business rules, describing user interfaces, providing web services and more.
Technical Overview of the platform is shown below.
|Data Processing Pipelines
|Templates, UI Components
SPARQL Web Pages (SWP), SWA
|Inference Engines, Data Mapping Rules
TopSPIN, BackSPIN, SPINMap
|Query Templates, Functions, Rules, Constraints
SPARQL Inferencing Notation (SPIN)
|Model: Classes, Properties
|Data Layer: Base Infrastructure, Databases
The foundation of TopBraid platform is RDF, which also serves as the base infrastructure of the Semantic Web and Linked Data. The platform implements W3C standards including RDF, RDFS, OWL and SPARQL and provides the following capabilities:
- Data Layer uses RDF to store and access information. Since most information today is stored in other formats, TopBraid dynamically accesses diverse data sources as if they were in RDF. This is possible because RDF data model is highly canonical and can represent any other data model including relational, XML, spreadsheet data, etc. The platform includes adapters for all popular data sources. TopBraid workspace is a named graph RDF store or 'triplestore'. This means that it has an important feature of organizing RDF triples into sets or graphs. Graphs can be persisted as files in TopBraid workspace or stored in the databases external to the workspace. TopBraid platform includes both an RDF-native and RDBMS-based database system for storing triples.
- Change Management layer maintains a history of all changes in the data.
- Query Engine uses SPARQL to access data.
- Models of information are represented using standard languages for semantic models – RDFS and OWL. These languages are build on RDF. This means that models are stored and can be queried the same way as data.
- Query Templates, Functions, Rules and Constraints are all build in SPARQL. SPIN provides mechanisms to attach behavior (rules and constraints) to classes in the model. It also enables parameterized SPARQL queries (query templates) that can be provisioned as web services. Furthermore, it makes it possible to create new custom SPARQL elements as functions. A key aspect of SPIN is that it allows users to define these rules and extensions in a completely declarative manner, entirely in RDF. This means that anything that is defined with SPIN can be shared and discovered in the same infrastructure that also holds the data.
- Inference and Data Mapping are provided through forward and backward chaining rules engines and SPIN-based language for defining mappings between models (SPINMap).
- Data Processing Pipelines are supported using SPARQLMotion, a declarative language with a visual notation and a corresponding engine. SPARQLMotion scripts consist of simple steps such as loading data, running transformations and then exporting the results. The result of one step may be passed as input to the next step, enabling extensive data transformation, alignment and integration capabilities. SPARQLMotion is aimed at modelers and business analysts who want to use high level graphical tools. SPARQLMotion can be used to define (web) services, and to create new SPARQL functions, which can then be used as part of other SPIN features.
- Templates and User Interface Components, as everything in the platform, are implemented using RDF. SPARQL Web Pages make it possible to define user interfaces by annotating semantic models with appropriate visualizations. SWP makes it easy to encapsulate HTML snippets into reusable components. TopBraid provides SWP Application Component Library (SWA) with pre-built trees, forms, grids components as well as more sophisticated visualizations.
Core Technologies and Application Development Tools