The correct answers are A. CSV stores flat, tabular data and E. JSON can be loaded into a VARIANT column without predefined schema .
Important correction:
The provided answer A, B is not correct. CSV does not natively support complex semi-structured data types such as arrays and objects. JSON does.
Why A is correct:
CSV is a delimited, flat file format commonly used for structured, tabular data. Each row usually represents a record, and each field is separated by a delimiter such as a comma.
Why E is correct:
JSON is semi-structured data and can be loaded into a Snowflake VARIANT column. This allows Snowflake to store JSON objects and arrays without requiring every nested field to be predefined as a relational column.
Example:
CREATE TABLE json_data (
src VARIANT
);
COPY INTO json_data
FROM @my_stage
FILE_FORMAT = (TYPE = JSON);
Why the other options are incorrect:
B). CSV does not natively support arrays and objects. JSON supports nested structures such as arrays and objects.
C). JSON does not need to be converted to CSV before it can be queried in Snowflake. Snowflake can query JSON data stored in VARIANT, OBJECT, or ARRAY columns.
D). CSV files do not automatically detect and apply Snowflake table data types during loading. The target table schema and file format options control how data is interpreted.
Official Snowflake documentation reference:
Snowflake documentation explains that CSV is used for structured data, while semi-structured formats such as JSON can be loaded into VARIANT columns and queried using semi-structured data syntax.
Reference: Snowflake Documentation - Semi-structured data; Snowflake Documentation - Loading JSON data; Snowflake Documentation - CSV file format options; SnowPro Core Study Guide - Data Loading and Unloading.
==

To start, resume, or suspend a warehouse, the role needs the OPERATE privilege. USAGE only allows reference to the object but not its activation.
Required Privileges for Starting a Warehouse:
* USAGE (to reference)
* OPERATE (to start/stop)
Reference:
Snowflake Docs: Warehouse Privileges

When a Snowflake virtual warehouse is resized, the data cached in the warehouse is not lost. This is because the cache is maintained independently of the warehouse size. Resizing a warehouse, whether scaling up or down, does not affect the cached data, ensuring that query performance is not impacted by such changes.
References:
[ COF-C03 ] SnowPro Core Certification Exam Study Guide
Snowflake Documentation onVirtual Warehouse Performance1