Integration Architecture Pattern and Principles

As set out in the Pathway to the Information Management Framework, the Integration Architecture is one of the three key technical components of the Information Management Framework, along with the Reference Data Library and the Foundation Data Model. It consists of the protocols that will enable the managed sharing of data across the National Digital Twin.

In the Integration Architecture Pattern and Principles paper, the National Digital Twin programme’s (NDTp) technical team sets out key architectural principles and functional components for the creation of this critical technical component. The team defines a redeployable architectural pattern that allows the publication, protection, discovery, query and retrieval of data that conforms to the NDT’s ecosystem of Reference Data Libraries and the NDT’s Foundation Data Model. 

The paper will take you through:

Recommended integration architecture pattern:

Three general architectural pattern options are explored in the paper (centralised, distributed, and federated). The benefits and concerns for each pattern are discussed with respect to the requirements. The recommended architectural pattern is a hybrid of these three approaches – centralising certain functions, whilst distributing and federating others. 

The recommended pattern is intended to allow datasets to be shared locally (i.e., within an NDT Node, see figure below), but will also allow for inter-node discovery, authorisation and data sharing to take place. NDT Nodes may be established by individual organisations, regulators and industry associations, or service providers and will be able to handle Digital Twins on behalf of their constituent organisations and provide a secure sharing boundary.

In the recommended architecture: Datasets are published by the data owner (1), these are then made available to the organisations within the community of interest, in addition an event is issued to register publication with the Core (2). When queries are submitted (A), the dataset can then be discovered by organisations in other communities of interest (B) and retrieved where appropriate (C). The release, discovery and retrieval are carried out according to the authorisation service so that access is controlled as specified by the data owner.

Detail of the functional components:

The Core Services are likely to be quite thin, comprising mainly of:

• a master NDT Catalogue that holds the location of available NDT Datasets across the ecosystem  
• the master FDM/RDL that will synchronise with the subset that is relevant for each NDT Node  
• a publish/ subscribe model to propagate data changes to parties that have an interest and appropriate contract in place.

The Core and each NDT Node shall interact through a microservice layer, with which they shall have to be compliant.

Next steps:The paper concludes with a list of ten key tasks to develop further the Integration Architecture components. We will make sure to inform you on progress and in the meantime, we are looking forward to hearing your questions and comments.