Madrid Instruments Service

This is a REST service that will simulate the control electronic instruments used in superconducting based quantum computers. The simulation is pretty straightforward, and not Quantum at all.

Service documentation

Endpoints

This is the endpoint for loading a program:

POST /program/load

The input of the service is a sequence of pulses and values that will map to some simple algebraic operations: Summation, Multiplication and Division.

The format of the input JSON should follow this schema:

{
   "program_code":[
      10,
      "Madrid_initial_state_pulse",
      120,
      "Madrid_pulse_1",
      3,
      "Madrid_pulse_2",
      "Madrid_pulse_2",
      2,
      "Madrid_pulse_2",
      "Madrid_pulse_1"
   ]
}

In case of success, the HTTP code will be 200 and the response JSON will be:

{
     "program_id": "MadridProgramId1"
}

This is the endpoint for triggering the execution of the program:

GET /program/run/<program_id>

In case of success, the HTTP code will be 200 and the response JSON will be:

{
     "result": 195
}

Workflow

Due to some hardware imposed constrains, we always need to load the program first, and then trigger the execution. When we call the /load_program endpoint with the corresponding request body and we will receive a response with a "Program ID". This "Program ID" needs to be used in a later call to the /run_program endpoint so the service identifies the program to run.

This implies two REST calls:

POST /program/load {...}
GET /program/run/<program_id>

Installing and running the application

The application is provided as a fastapi application, which can be run using uvicorn (included as a dependency for convenience) or any other ASGI-server.

Installing the application

$ pip install .

This will install the application and all its dependencies.

Executing the application

$ uvicorn madrid_instruments_service.main:app
INFO:     Started server process [477295]
INFO:     Waiting for application startup.
INFO:     Application startup complete.
INFO:     Uvicorn running on http://127.0.0.1:8000 (Press CTRL+C to quit)

This will execute the application, listening by default at http://127.0.0.1:8000. You can also check the documentation for the endpoints at http://127.0.0.1:8000/docs.

atilag / MadridInstrumentsService