DuckDB

DuckDB (opens in a new tab) allows you to use Glean without setting up any external database.

DuckDB is a unique database amongst the types Glean supports. Traditional analytical databases are optimized to store massive quantities of data that can be retrieved at high speeds. While these databases are incredibly powerful, they often require a fair amount of work to set up.

DuckDB is instead focused on getting you up and running with analyzing a (smaller) dataset as quickly as possible. It can read common file types you likely already have: CSVs, JSON arrays, and more. DuckDB also runs on Glean's servers, so you don't need to set up (or pay for) any infrastructure to start querying a dataset.

This makes DuckDB perfect for one-off analysis, or for less technical users to get their data into Glean without setting up any servers. DuckDB is not a good choice if your dataset is large, or needs to be updated in realtime from an external system.

How to get set up

Glean pre-configures a DuckDB connection for every project. You are already ready to start exploring data files in your Glean project.

See Modeling with Uploaded Files for more information.

Organizing file groups

If you have lots of uploaded files, it may be helpful to organize them into distinct groups. To do so, you can create a new data connection of the DuckDB type.

1. First, go to your Glean settings (opens in a new tab) page from the project dropdown.
2. Click + New Database Connection and select Uploaded Files (DuckDB).