Firebolt

Firebolt is a cloud data warehouse built for low-latency analytics. Queries run on a named engine (a compute cluster) against a database, and both are named in the request rather than in a hostname. AddisDB submits SQL over its HTTP query API with a bearer token.

Who it is for

Firebolt targets interactive, user-facing analytics — dashboards and in-product reporting where a multi-second query is too slow. Its indexing model (aggregating and join indexes) is the main lever, and the main thing to learn.

The operational detail that matters most: an engine must be RUNNING to answer queries. A stopped engine refuses them, and that refusal reads like an authentication problem if you are not expecting it.

Set up the service account

1. In the Firebolt console create a service account and note its client ID and secret.
2. Exchange those for an access token against Firebolt’s identity endpoint. AddisDB takes the token, not the client secret — the same posture as Databricks and Dremio here.
3. Note your account endpoint, which looks like my-account.us-east-1.firebolt.io.
4. Note the database, and the engine you want queries to run on. Start the engine if it is stopped.

curl -X POST https://id.us-east-1.app.firebolt.io/oauth/token \
  -d "grant_type=client_credentials" \
  -d "client_id=$ID" -d "client_secret=$SECRET"

Connect from AddisDB

1. New Connection → Firebolt.
2. Put the account endpoint in the host field.
3. Put the database in the database field, as database or database/engine.
4. Paste the access token into the password field.
5. Test, then Save.

What AddisDB gives you

Queries and the Big Data Console, the schema tree from information_schema, exports, saved queries and mock data.

Firebolt reports query errors inside a successful HTTP response. AddisDB checks the response body rather than the status code, so a syntax error arrives as Firebolt’s own message instead of an empty grid.