---
tags: Legacy
---

# Docker Integration (Revenue Sharing)

## Functionality Requirements

- Run simulations in the container
- Fire up jupyter notebook for coding from the container

## Design Questions

- Should volumes be used to persist data
- Should we have two different docker containers, one for jupyter, one for running simulations?
- Could we do a base image python or miniconda container, create two containers for these two, then use docker compose to allow for running both?

## Resources

- [Jupyter Implementation](https://jupyter-docker-stacks.readthedocs.io/en/latest/)

## Proposed Plan

- Base image of python or miniconda
- Two images, one for jupyter + one for simple running that are built on the base image
- Use docker compose and allow for both to be run
- Jupyter notebook will be launched on the port 8888
- Other will need an interface to allow for sending commands like testing or running large scale simulations