Together with Zorg van de Zaak, AMIS Conclusion has drawn up an approach in 4 phases: inventory, blueprint, infrastructure and implementation. This ensures a structured implementation while keeping the project fully manageable.
Taking inventory
In the inventory phase, we mapped out the data sources and data customers together with Zorg van de Zaak. We record which internal and external data sources there are, in what format and with what frequency they deliver data. For the data buyers, we have mapped out which data they use, how it is presented and with what frequency. We also took into account future plans, such as new data sources or dashboards.Zorg van de Zaak works with medical data and therefore falls into the highest security and privacy category. To ensure this, we have applied security and privacy-by-design. We already provide this classification for the source system, so that the same privacy and security rules apply throughout the system. This makes working with the data safe, reliable and traceable.
Blueprint
In the blueprint phase, we inventoried the necessary cloud components and figured out how the on-premise environment connects to the cloud environment. The result of this phase is a design of the new infrastructure.
Infrastructure
In de infrastructuurfase hebben we de volledige omgevingen opgezet als Infrastructure As Code (IAC). IAC maakt het mogelijk om identieke omgevingen op te leveren op basis van code en indien nodig een extra omgeving op te leveren. Dit was het fundament voor de volgende fase.
Implementation
In the implementation phase, we built pipelines for each data source that ensure that this data becomes available as data products in the cloud. We eventually used these data products for the reports and dashboards. A data pipeline describes the actions to move data from source system to target system. To keep our finger on the pulse, we have set up monitoring so that we know whether everything is working properly.
For the solution, AMIS Conclusion used various Azure components, including Azure Functions, Azure Data Lake and Azure Synapse