Datomic Analytics in Docker container

Datomic Analytics uses an older version of Presto (now called Trino) which has some limitations:

• doesn't run on Apple Silicon
• requires Java 11

These instructions show how to get Datomic Analytics running in a docker container, and connect it to a Datomic Cloud instance. For Datomic On-Prem you might have to do some additional tweaks, according to Datomic Analytics docs.

The reason we made this was to get a Datomic Analytics/Trino server running on Apple Silicon on a development machine. You should be able to use this for production deployments, by editing the files in the etc directory.

Datomic Analytics docs

Read the Datomic Analytics documentation.

Clone repo

Clone this repo to get a local copy of all files.

Configure Datomic Analytics/Trino

Edit or replace these files with the correct files for your setup:

• catalog: etc/catalog/catalog.properties
• metaschema: etc/datomic/mbrainz.edn

The catalog.properties included here contains a sample config to connect to a Datomic Cloud instance. Replace the placeholders with values pertaining to your setup.

Build docker image

From the root folder of this 'project', run:

docker build . -t "datomic/trino" -f ./Dockerfile

You can replace "datomic/trino" with any name of your choice, but if you do be sure to use the same value in the next step.

Run the image in Docker

When connecting Trino to a Datomic Cloud system, you will need your AWS credentials (aws_access_key_id and aws_secret_access_key) for the next step.

docker run --name datomic-trino -d -p 8989:8989 datomic/trino [aws_access_key_id] [aws_secret_access_key]

Trino should now be running and available on port 8989 on your local machine.

Trino CLI

Follow the instructions in Datomic docs to install the Trino CLI and connect to your Trino instance.