Apache Doris

Apache Doris is a massively-parallel columnar database for real-time analytics. Its front-end nodes speak the MySQL wire protocol, so AddisDB connects with its MySQL driver on port 9030. Doris and StarRocks share an ancestor, which is why they look so alike from the outside.

Who it is for

Doris suits reporting and interactive analytics where data arrives continuously and queries need to see it quickly — it ingests from Kafka, Flink and object storage and answers aggregate queries over large tables without a separate serving layer.

As with StarRocks, the table model is chosen at creation. Duplicate keys store rows as-is, aggregate keys roll them up on write, and unique keys give you upserts. It is worth knowing which model a table uses before wondering why a row you inserted looks different when you read it back.

Set up the cluster

1. Note a front-end (FE) host — any FE serves queries.
2. Use the MySQL protocol port, 9030. The FE HTTP port 8030 serves the web UI and is a different thing.
3. The initial account is root with an empty password; set one before the cluster is reachable from anywhere but localhost.
4. Note the database you want to browse.

mysql -h doris-fe.example.com -P 9030 -u root -e 'SHOW DATABASES;'

Connect from AddisDB

1. New Connection → Apache Doris. Port prefills to 9030 and username to root.
2. Enter the FE host and credentials, and the database name.
3. Set SSL mode to require if the cluster terminates TLS.
4. Test, then Save.

What AddisDB gives you

The Big Data Console, since Doris is catalogued as a warehouse engine, plus the results grid, exports and saved queries.

The schema tree reads information_schema, which Doris implements, so databases, tables and columns load normally.

Query plans use Doris’s own EXPLAIN and render as text — the planner’s estimate, with no measured per-node timings.