Reference Architecture Consent Management in Data Spaces

This project was part of my Master Thesis project for the UvA Software Engineering Computer Science Master degree executed at TNO.

This application works on any Kubernetes cluster (such as a local one), as well as Cloud based Kubernetes Engines.

Service Architecture

The reference architecture is composed of many components (micro-services) written in different languages that talk to each other over REST.

Current Functionality

Data Provider

Registration of data sets
Registration of data owners
Listing of data sets
Listing of data owners
Registration of data sets and data owners at a consent manager

Consent Manager

Registration of data sets
Registration of data owners
Registration of data providers
Definition of data permissions
Definition of data obligations
Linking of permissions and obligations with data sets
Registering policies at the policy catalogue

Policy Catalogue

Registration of policies
Listing of policies
Deleting of policies
Retrieving of policies (the policy content)

Policy Broker

Registration of consent managers
Registration of data categories belonging to consent managers
Registration of data permissions and obligations of consent managers

Features

Kubernetes: The infrastructure is designed to run on Kubernetes (both locally on "Docker for Desktop", as well as on the cloud).
Flask: Public microservices make use of flask to provide a REST interface.
Skaffold: Applications are deployed to Kubernetes with a single command using Skaffold.

Installation

Installation of the infrastructure kan be done locally and in the cloud.

The following installation methods are supported:

Running locally (~20 minutes)

You will build and deploy the microservices images to a single-node Kubernetes cluster running on your development machine. You need install the following requirements:

Prerequisites

Docker. For containerization of the applications.
Kubectl. To interact with your kubernetes cluster.
Minikube. Recommended for Linux hosts (also supports Mac/Windows).
Virtualbox. Used as a driver for minikube.
skaffold (ensure version ≥v1.10)

Start the minikube cluster

In the root directory run the following command:
```
sh setup-minikube.sh
```
Deploy the database infrastructure

First the deploy the postgres operator
```
cd databases_infrastructure
sh setup_dev_environment.sh
```
Wait till the postgres operator is online. You can check this by running the following command:
```
kubectl get pods
```
The 'status' of the postgres operator should have the value running
```
NAME                                 READY   STATUS    RESTARTS   AGE
postgres-operator-85978df8cf-kgb9z   1/1     Running   0          7s
```
When the operator is running, you can deploy the database instances with the following command:
```
kubectl apply -f manifests/services-cluster-deployment.yaml
```
Deploy the service instracture

Starting from the root directory run the following script in the service_infrastructure directory.
```
cd services_infrastructure
sh setup_skaffold.sh
```
After all the images are deployed you can update your host file with the following command:
```
cd ..
sh manage-hosts.sh
```
Deploy the frontend

(As of now the frontend is not yet integrated in the kubernetes cluster. Therefore it is run locally.)

You can deploy the frontend by going to the frontend service directory and run the following commands:
```
cd services_infrastructure/frontend
npm init
npm run dev
```
Your frontend should then be available at: http://localhost:3000

MDUYN / reference-architecture-consent-management-in-data-spaces