COF-C03 Exam Dumps - SnowPro Core Certification 2026 Exam

The correct answer is C. SAML .

In Snowflake, Single Sign-On is commonly implemented using federated authentication through a SAML 2.0 identity provider. When creating an authentication policy, including SAML in the allowed AUTHENTICATION_METHODS enables SSO-based login options.

Why C is correct:

SAML represents SAML-based federated authentication. This is the authentication method associated with SSO login in Snowflake.

Example concept:

CREATE AUTHENTICATION POLICY sso_policy

AUTHENTICATION_METHODS = ( ' SAML ' );

Why the other options are incorrect:

A. PASSWORD allows username and password authentication, not SSO.

B. KEYPAIR allows key-pair authentication, commonly used for service accounts and applications.

D. OAUTH allows OAuth-based authentication, but the SSO login option in this context is associated with SAML.

Official Snowflake documentation reference:

Snowflake documentation describes authentication policies as controlling allowed authentication methods, including password, SAML, OAuth, and key-pair authentication. SAML is used for federated SSO authentication.

The correct answers are C. Metadata management and E. Query parsing .

Snowflake’s architecture has three main layers:

Cloud Services Layer

Query Processing Layer / Compute Layer

Database Storage Layer

The cloud services layer coordinates Snowflake activities and handles services such as authentication, infrastructure management, metadata management, query parsing and optimization, access control, and transaction management.

Why C is correct:

Metadata management is handled by the cloud services layer. This includes metadata about tables, micro-partitions, query optimization statistics, and access-related metadata.

Why E is correct:

Query parsing and optimization are handled by the cloud services layer before execution is assigned to compute resources.

Why the other options are incorrect:

A. Performance tuning is not usually described as a discrete cloud services task. Snowflake automatically optimizes many operations, but this option is too broad.

B. Query acceleration uses additional compute resources and is a performance feature, not a core task of the cloud services layer.

D. Query processing is performed by virtual warehouses in the compute layer, not by the cloud services layer.

Official Snowflake documentation reference:

Snowflake documentation explains that the cloud services layer manages authentication, infrastructure management, metadata, query parsing and optimization, access control, and transaction management.

The correct answer is D. File-level metadata of all data files in an internal or external stage .

A Snowflake directory table is associated with a stage and stores file-level metadata for files in that stage. It lets users query staged file information using SQL.

Why D is correct:

Directory tables hold metadata about staged files, such as file name, relative path, size, last modified time, and file URL information.

Example concept:

SELECT *

FROM DIRECTORY(@my_stage);

This returns metadata for files in the stage associated with the directory table.

Why the other options are incorrect:

A. Directory tables do not store metadata about views in a schema.

B. Directory tables do not store metadata about regular Snowflake tables in a schema.

C. Directory tables do not store clustering key metadata.

Official Snowflake documentation reference:

Snowflake documentation describes directory tables as objects that store file-level metadata about staged files in cloud storage.

The correct answer is C. A role with the OWNERSHIP privilege on the database .

A database role is scoped to a specific database. To create a database role, a role must have sufficient privileges on the database where the database role will be created.

Why C is correct:

The role that owns the database has control over that database and can create database roles within it.

Example:

CREATE DATABASE ROLE my_database.my_db_role;

To run this successfully, the executing role must have the required privilege on the database, such as OWNERSHIP.

Why the other options are incorrect:

A. SECURITYADMIN manages users, roles, and grants at the account level, but database role creation depends on privileges on the specific database.

B. A custom role being granted to SECURITYADMIN does not automatically mean it can create database roles in a database.

D. SYSADMIN commonly owns many account objects by default in recommended role models, but the required condition is privilege on the database, not simply holding the SYSADMIN role.

Official Snowflake documentation reference:

Snowflake documentation explains that database roles are created inside databases and require appropriate privileges on the database.

To optimize parallelism and maximize performance, large files should be split into multiple smaller files (ideally 100–250 MB compressed) before loading. This allows Snowflake to load files in parallel using multiple threads or clusters.