Release Note 5.2.30

Client Communication

Updates and Changes to the User Experience in Spaces

April 2023

Presto has been updated and renamed Trino.

In the Exchange and Spaces user interface, the Presto tool has been upgraded and renamed “Trino”. It is now the default query engine in Spaces.

The new Trino Spaces tool has the following logo:

Execute Faster Queries with Trino

Trino is an open-source distributed query engine that is highly optimized for query performance across multiple data sources and is visible when you Activate your Tools and Compute. For Spaces this means faster performance and a secured connection to Snowflake to execute queries against Assets that have been added to a Space.

For guidance on using Trino within Spaces see Using SQL in a Space, Using Python in a Space and Combine Data Assets and Data Products in a Space.

Updated Connections and Table Names

The HUE editor allowing access to the Data Product, Collaborate and Publish databases is now renamed “Spaces” replacing “PrestoSQL”.

In Superset SQLLab, Trino will be provisioned as the ‘Space’ database, replacing ‘Fast Spaces Metastore.’

In pre-existing Spaces, the ‘Fast Spaces Metastore’ will initially be maintained for compatibility but may be removed in a later release.

The underlying Trino catalog that supports these connections will be called ‘spaces’ rather than ‘hive’, such that fully qualified table names in queries would change:
‘hive.collaborate_db.output_table1’ (Presto) → becomes → ‘spaces.collaborate_db.output_table1’ (Trino)

Within JupyterLab notebooks, PyHive connections should be updated with the following lines (all subsequent python actions remain the same):

• from pyhive import trino

• conn = trino.Connection(host=”localhost”, port = 8060)

Within RStudio, the RPresto package should continue to be used with the following line added to the DBI::dbConnect() connection :

• use.trino.headers = True