Splitgraph has been acquired by EDB! Read the blog post.

Database structure

A fresh Seafowl installation has a single database, default. Seafowl doesn't support creating or connecting to a different database.

public schema

public is the default schema for all Seafowl tables. You do not need to prefix table references with a schema name if they are in the public schema.

To create a different schema, use the CREATE SCHEMA statement.

staging schema

This is a dedicated temporary schema for external (live) tables. You can create tables in this schema with, and only with, the CREATE EXTERNAL TABLE statement. Tables in this schema don't get persisted into Seafowl and don't survive instance restarts.

system schema

This schema contains helper tables with the aim of making Seafowl's internal state more transparent. For now the only tables available are table_versions and dropped_tables; they can be used to query the available table versions (including creation timestamps) as well as tables that were dropped and need to be VACUUM-ed.


Seafowl exposes DataFusion's information_schema that provides information about the available tables and their schemas:

  • information_schema.tables: list of all tables
  • information_schema.columns: list of all table columns and their types
  • information_schema.views: list of all views

See DataFusion's information_schema documentation for more information.