PHT Architecture Components
The Tübingen implementation of the PHT has several central services which are used to submit, control and execute trains. Our current implementation relies on the manual acceptance of each train before it can be executed at a hospital. Details regarding our implementation are described below.
Overall worflow
- User is redirected to associated hospital local IAM
- Send Auth Code to UI
- Exchange Auth code for access token
- Sign hash from files during train submission
- UI sends building message to Message Broker
- Train Building receives message queue
- Post route of train to Vault
- Receive public keys of stations
- Build train and push in incoming repository
- Send event for incoming train
- Publish train event to Train Routing
- Receive route of train in Vault
- Move trains according to route and execution status
- Pull image
- Execute train at station locally
- Push executed train with encrypted results
- Result extraction of executed trains
- Return encrypted results
- Decrypt results locally with Desktop App
Central User Interface and train submission
Execution at stations
The execution of a train is separated in 3 Phases:
- Pre-run protocol
- Container execution
- Post-run protocol
During first phase a validation check for manipulation of the algorithms and decryption of the model with envelope encryption is made. The run phase executes the train image including all software requirements wrapped within the container to run the algorithm at one station. The third and last phase encrypts the train results and prepares the train for the departure. Only encrypted results are hosted centrally and must be decrypted locally. The execution (currently using the airflow interface) of a train can be seen in the video on the right.
