The TPCH connector provides a set of schemas to support the TPC Benchmark™ H (TPC-H). TPC-H is a database benchmark used to measure the performance of highly-complex decision support databases.
This connector can be used to test the capabilities and query syntax of Trino without configuring access to an external data source. When you query a TPCH schema, the connector generates the data on the fly using a deterministic algorithm.
To configure the TPCH connector, create a catalog properties file
etc/catalog/example.properties with the following contents:
In the TPC-H specification, each column is assigned a prefix based on its
corresponding table name, such as
l_ for the
lineitem table. By default, the
TPCH connector simplifies column names by excluding these prefixes with the
SIMPLIFIED. To use the long, standard
column names, use the configuration in the catalog properties file:
The TPCH connector supplies several schemas:
SHOW SCHEMAS FROM example;
Ignore the standard schema
information_schema, which exists in every
catalog, and is not directly provided by the TPCH connector.
Every TPCH schema provides the same set of tables. Some tables are
identical in all schemas. Other tables vary based on the scale factor,
which is determined based on the schema name. For example, the schema
sf1 corresponds to scale factor
1 and the schema
corresponds to scale factor
300. The TPCH connector provides an
infinite number of schemas for any scale factor, not just the few common
ones listed by
SHOW SCHEMAS. The
tiny schema is an alias for scale
0.01, which is a very small data set useful for testing.
Trino supports all data types used within the TPCH schemas so no mapping is required.