Skip to main content

Snowflake

Dozer's Snowflake connector integration utilizes Snowflake's Table Streams for real-time data integration and synchronization. Dozer captures and processes data changes from Snowflake tables using Table Streams, ensuring instant updates without constant polling or manual synchronization.

Configuration

The following configuration block can be used in dozer-config.yaml to define a new Snowflake connection:

connections:
- config: !Snowflake
server: "dozer-test.snowflakecomputing.com"
port: 443
user: "dozerTest"
password: "DozerSamplePassword!"
database: "DOZER_SAMPLE_DATA"
schema: PUBLIC
warehouse: "COMPUTE_WH"
driver: "SnowflakeDSIIDriver"
role: "ACCOUNTADMIN"

Parameters

Parameter NameTypeDescription
serverStringThe server URL or endpoint required for authenticating and connecting to the user's Snowflake account.
portIntegerThe specific port on which the Snowflake service is running. By default, Snowflake uses port 443 for encrypted communication.
userStringThe username required for authenticating the user's access to the Snowflake instance.
passwordStringThe password corresponding to the above username, required for secure authentication to the Snowflake instance.
databaseStringThe specific database within the Snowflake instance to which the connector needs to establish a connection. This is the database that will be used as the default for the session upon successful connection.
schemaStringThe specific schema within the selected database that the connector needs to establish a connection. A schema is a logical container for database objects like tables, views, and procedures.
warehouseStringThe name of the warehouse to which the connector needs to connect. This determines the virtual compute resources available for executing queries and processing data within Snowflake.
driverStringThe type of driver or connector used to establish the connection with Snowflake.
roleStringThe role to which the user is assigned within Snowflake.

Trying it out

To test a Snowflake sample, clone the dozer-samples GitHub repo and follow the steps described here.