This project provides a simple setup for a Grafana installation, a PostgreSQL database and a pluggable set of data loaders that load data in the db, which in turns is the main data source for Grafana.

For example, garmin-importer provides a way to export sports activities from Garmin Connect and into an own storage of choice (in that case, dynamo-db), and dashed is where that data is put to good use.

The data loader concept and implementation should be generic enough that it should be easy to replicate, so that other data can be imported from other sources. This will make it possible to extend the reach of our insights once we can cross-reference data from multiple sources.

It is an industry standard, open-source, platform that gives users the possibility of crafting their own dashboards.

We will run the latest stable release from its docker container, and find a way to persist dashboards in a way that if we re-start this whole project on separate hardware, we get the desired set of dashboards up and running.

This is also an open-source industry standard and Grafana has a stable plugin-integration with it

We will run a single instance docker container based on the latest stable release. Each data loader will take care of making sure that the necessary tables are set up, that there are no migrations to run, and each data loader should create its own dedicated user to insert data and to create and manage its own table, so that different loaders won't be able to interfere with each other.

This is the component that at startup will check what data is in the database, make sure that the tables it needs are there and with the correct schema, that all the db migrations have been carried out, and import the necessary data.