Snowflake reorganizes data into its internal optimized, compressed, columnar format in the

Database Storage layer. This process is part of how Snowflake manages data storage, ensuring

efficient data retrieval and query performance

Once a database has been created in a consumer account from a share, the shared objects become

accessible to users in that account. The shared objects are not transferred or copied; they remain in

the provider’s account and are accessible to the consumer account

In Snowflake, to create a materialized view, the user must have the necessary privileges on the

schema where the view will be created. These privileges are granted through roles, not directly to

individual users. Therefore, the correct process is to grant the role to the user and then grant the

privilege to create the materialized view to the role itself.

The statement GRANT ROLE MYROLE TO USER USER1; grants the specified role to the user, allowing

them to assume that role and exercise its privileges. The subsequent statement CREATE

MATERIALIZED VIEW ON SCHEMA MYDB.MYSCHEMA TO MYROLE; grants the privilege to create a

materialized view within the specified schema to the role MYROLE. Any user who has been

granted MYROLE can then create materialized views in MYDB.MYSCHEMA.

Reference:

Snowflake Documentation on Roles

Snowflake Documentation on Materialized Views

Snowflake’s secure data sharing feature allows sharing of certain database objects with other

Snowflake accounts. Among the options provided, external tables and secure UDFs can be shared

Snowflake supports several methods for authenticating users, including federated

authentication, key-pair authentication, and OAuth. Federated authentication allows users to

authenticate using their organization’s identity provider. Key-pair authentication uses a public-

private key pair for secure login, and OAuth is an open standard for access delegation commonly

used for token-based authentication. Reference: Authentication policies | Snowflake

Documentation, Authenticating to the server | Snowflake Documentation, External API

authentication and secrets | Snowflake Documentation.

The cloud services layer of Snowflake architecture handles various aspects including security

functions, authentication of user sessions, and access control, ensuring that only authorized users

can access the data and services23.

The UNDROP command in Snowflake can be used to directly restore Views and Tables. These objects,

when dropped, are moved to a ‘Recycle Bin’ where they can be restored within a time limit before

they are permanently deleted. Reference: [COF-C02] SnowPro Core Certification Exam Study Guide

A Pre-signed URL provides access to files stored in Snowflake without requiring authorization at the

time of access. This feature allows users to generate a URL with a limited validity period that grants

temporary access to a file in a secure manner. It's particularly useful for sharing data with external

parties or applications without the need for them to authenticate directly with Snowflake.

Reference:

Snowflake Documentation: Using Pre-signed URLs

Virtual Warehouses in Snowflake can indeed be resized while they are suspended. Resizing a

warehouse involves changing the number of compute resources (servers) allocated to it, which can

be done to adjust performance and cost. When a warehouse is suspended, it is not currently running

any queries, but its definition and metadata remain intact, allowing for modifications like resizing.

Reference: https://docs.snowflake.com/en/user-guide/warehouses-tasks.html#effects-of-resizing-asuspended-warehouse

Snowflake does not require source data files to be no larger than 16MB. In fact, Snowflake

recommends that for optimal load performance, data files should be roughly 100-250 MB in size

when compressed. However, it is not recommended to load very large files (e.g., 100 GB or larger)

due to potential delays and wasted credits if errors occur. Smaller files should be aggregated to

minimize processing overhead, and larger files should be split to distribute the load among compute

resources in an active warehouse.

Reference: Preparing your data files | Snowflake Documentation