In Snowflake, the consumption of credits by a virtual warehouse is determined by its size and the duration for which it runs. A size 3X-Large virtual warehouse consumes 128 credits if it runs continuously for 2 hours.
This consumption rate is based on the principle that larger warehouses, capable of providing greater computational resources and throughput, consume more credits per hour of operation. The specific rate of consumption is defined by Snowflake's pricing model and the scale of the virtual warehouse.References:Snowflake Pricing Documentation

The PARTITION BY <expression> parameter option in the COPY INTO <location> command is used to split the output into multiple files based on the distinct values of the specified expression. This feature is particularly useful for organizing large datasets into smaller, more manageable files and can help with optimizing downstream processing or consumption of the data. For example, if you are unloading a large dataset of transactions and use PARTITION BY DATE(transactions.transaction_date), Snowflake generates a separate output file for each unique transaction date, facilitating easier data management and access.
This approach to data unloading can significantly improve efficiency when dealing with large volumes of data by enabling parallel processing and simplifying data retrieval based on specific criteria or dimensions.
Reference:
Snowflake Documentation on Unloading Data: COPY INTO <location>

In Snowflake, the consumption of credits by a virtual warehouse is determined by its size and the duration for which it runs. A size 3X-Large virtual warehouse consumes 128 credits if it runs continuously for 2 hours. This consumption rate is based on the principle that larger warehouses, capable of providing greater computational resources and throughput, consume more credits per hour of operation. The specific rate of consumption is defined by Snowflake's pricing model and the scale of the virtual warehouse.
Reference: Snowflake Pricing Documentation