Semi-structured
JSON, Parquet
Snowflake supports unloading data in several semi-structured file formats, including Parquet and JSON. These formats allow for efficient storage and querying of semi-structured data, which can be loaded directly into Snowflake tables without requiring a predefined schema12.
https://docs.snowflake.com/en/user-guide/data-unload-prepare.html#:~:text=Supported%20File%20Formats,-The%20following%20file&text=Delimited%20(CSV%2C%20TSV%2C%20etc.)

* Tasks in Snowflake are used to create workflows or schedules for executing SQL statements, including transformations, in a specific order.
* By defining dependencies between tasks, users can ensure they execute in a defined sequence.
* Example:
sql
CopyEdit
CREATE TASK task_1
SCHEDULE = '5 MINUTE'
AS
INSERT INTO transformed_table SELECT * FROM raw_table;
CREATE TASK task_2
AFTER task_1
AS
DELETE FROM raw_table WHERE processed = TRUE;
* Tasks can also run on a fixed schedule or be triggered by preceding tasks in a chain.
Why Other Options Are Incorrect:
* A. Pipes: Automates data loading but does not handle transformation workflows.
* C. Streams: Tracks table changes but does not schedule or sequence transformations.
* D. Sequences: Generate unique numbers, unrelated to task scheduling or transformations.
References:
* Snowflake Tasks Documentation