Snowflake Table Destination

The Snowflake Table destination writes data to a single Snowflake table. To write to more than one table, use additional destinations.

When you configure the destination, you specify the table to write to and whether the destination should create the table if it does not exist. You define the write mode and related properties. You can also enable data drift handling, which updates the target table when the pipeline rows contain unexpected data.

By default, the destination connects to Snowflake using the Snowflake pipeline properties. When needed, you can specify another database or schema to use instead.

Create Table

The Snowflake Table destination can create a table before writing to it. When you enable the Create Table property, the Snowflake Table destination creates a table based on the data to be written to the table. The destination then writes to the table based on the selected write mode and related properties.

On subsequent pipeline runs, the Snowflake Table destination writes to the table based on the write mode and related properties. It does not drop and recreate the table.

Write Mode

The write mode defines how the Snowflake Table destination writes to an existing table, including the second batch to a table that the destination created.

You can use one of the following write modes with the Snowflake destination:
Append to table
The destination appends rows to an existing table.
Use append mode only when each new row has a unique primary key. If a row with the same primary key exists in the table, the pipeline fails with a primary key violation.
Overwrite table
The destination removes all rows in the existing table before writing new rows to the table. When you use this option, you also specify the overwrite mode to use:
  • Drop Table - Drops the existing table, creates a new table based on pipeline data, and then writes to the table.
  • Truncate Table - Truncates the data in the existing table, then writes to the table.
Merge rows in table by condition
The destination merges data to an existing table based on the configured merge keys and other merge properties.
You can configure the destination to automatically merge data or specify merge conditions. For more information, see Merging Data.
Propagate updates from Slowly Changing Dimension processor
The destination inserts and updates rows provided by the Slowly Changing Dimension processor.
Use only in a pipeline that uses a Slowly Changing Dimension processor.

Merging Data

When you merge data, you specify one or more columns in the row to use as keys for the merge.
Important: The target table must have columns with matching names. The match is not case-sensitive.
After specifying the merge keys, you use one of the following merge options:
Autocalculate Merge
When the destination automatically calculates the merge, information in the following row columns determines how each row is written to the target table:
  • __SS_META_KEYS - List of columns to use as keys.
  • __SS_META_OPERATION - Operation to perform for the row. Uses the following values:
    • 1 - Insert
    • 2 - Update
    • 3 - Delete
Note: Before using the Autocalculate Merge option, ensure that the pipeline data includes these columns and that the columns are set appropriately. You can use the Snowflake SQL Evaluator processor to create and set the columns.
The destination removes the __SS_META_KEYS and __SS_META_OPERATION columns from the data before it is written to the target table.
Merge Configuration
When the destination uses merge configuration properties, you define one or more sets of configuration properties. Typically, you specify at least one WHEN MATCHED and one WHEN NOT MATCHED clause. Specify sets of configuration properties in the order that they should appear in the SQL query.
Each merge configuration includes the following properties:
  • When Clause - Create a WHEN MATCHED clause or a WHEN NOT MATCHED clause.
  • Match Operation:
    • When using a WHEN MATCHED clause, specify whether to update or delete rows upon matching the configured set of keys.
      Note: Unlike Snowflake, the Snowflake Table destination matches on null values.
    • When using a WHEN NOT MATCHED clause, specify whether to insert or ignore rows that have no matching set of keys.
  • Additional Not Matched Condition - Optionally specify an additional predicate for a WHEN NOT MATCHED merge condition.
For example, you might use the following configuration properties to update existing rows and insert all new rows:

Data Drift Handling

You can configure the Snowflake Table destination to handle data drift. Data drift is an unexpected change in the structure or format of the data, such as additional columns or column data changing from integer to decimal data.

When handling data drift, the destination updates the target table to enable writing data with these unexpected changes to the table. Enable data drift only when you want to enable data with unexpected shifts in data structures to be written to the target table.

Do not enable data drift when you do not expect shifts in data structure or types and do not want such rows to be written to the destination, or when you simply do not want the target table to be changed based on incoming data.

When you enable data drift, the destination performs the following tasks, as needed, to enable writing pipeline data to the table:
  • If the table does not exist, the destination creates the table
  • If pipeline data includes a column that the target table does not, the destination issues an ALTER TABLE ADD COLUMN statement to add all required columns.

    String columns are created as Varchar(n) where n is based on the maximum precision for the column in the first batch being written.

Note: When appending data to an existing table, if the target table has a column that the source table does not, the pipeline adds the column to the source table and populates the column with null values.

This ensures that the source and target table have the same structure.

Configuring a Snowflake Table Destination

Configure a Snowflake Table destination to write data to a single Snowflake table. To write to more than one table, use additional destinations.

  1. On the General tab, configure the following properties:
    General Property Description
    Name Stage name.
    Description Optional description.
  2. On the Table tab, configure the following properties:
    Table Property Description
    Table Table to write to. Use one of the following methods to define this property:
    • Manually enter - Click the Table text box. In the Overwrite Database, Schema, and Table dialog box, you can specify the database, schema, and the table or view to use, then click Save. The database or schema specified here overrides those defined in pipeline properties.
    • Explore and select - Click the Select Table icon to explore your Snowflake data and navigate to the table or view to use.
    Write Mode Write mode to use.
    Overwrite Mode Overwrite mode to use.

    Available when overwriting the table.

    Create Table Enables creating the specified table if it does not exist.
    Data Drift Enabled Automatically compensates for new or missing data in pipeline rows. Enables adding new columns to the table and populating missing columns with null values.