Data Centric Architecture

Owned by DevPortal Editor (Unlicensed)
Last updated: 23/03/22 by Onesait Platform
2 min read

 

INFO

From version 3.1.0 in the Control Panel, Ontologies have been renamed as Entities in the Control Panel. This does not alter any functionality, simply the nomenclature has been changed for a better understanding of the concept.

Intro

Data-centric refers to an architecture where data is the main, permanent asset, and applications come and go. In the Data-centric architecture, the data model precedes any given application's implementation and will be valid long after these are replaced.

Although it may seem like a typical architecture, the reality is that it rarely happens exactly that way. Companies looking for functionality buy or build applications. Each application has its own data model and its code is inextricably linked to that data model. It is extremely difficult to change the data model of an implemented system, since there may be millions of lines of code depending on this existing model. In addition, this application is just one of the hundreds or thousands of systems in a company. Each application itself has hundreds to thousands of tables and tens of thousands of attributes. These applications are interconnected in a very partial and very unstable way through a middleware that periodically transports data from one database to another.

The Data-centric approach turns all this into your core. There is a data model, a semantic data model and each application functionality reads and writes through the shared model. If there is an application functionality that calculates KPIs, it will add it to the shared database, using the common basic terms.

Any other system can access the KPIs and know what they mean. If the functionality disappears tomorrow, the KPIs will still be there.

Data-Centric: the Ontology

We support the Data-Centric architecture through the Ontology, and all the functionality of the Platform looks around this concept. The Ontologies are the Entities that the system manages. In the simplest case, we can compare an ontology with a table in a relational database, but an ontology can contain an entire domain model, which would require a set of related tables in a relational database.

The ontology can persist in different repositories and isolates the user from the underlying repository, so that the rest of the components do not need to know where the ontologies are.

The other concepts and components of the platform orbit around the Ontology include, among others, these:

  • The creator of the ontology will assign security of access on the ontology to the rest of users.

  • The ontology can be represented through the dashboards.

  • The rules are defined and associated with the arrival of a given ontology to the system.

  • APIs are created on an ontology.

  • ...

Functionality around the Ontology