COF-C03 - Data Loading, Unloading, and Connectivity - Section 3.1

Describe internal stages (user, table, and named) and external stages, and the file format options for loading and unloading.

Describe the stages that hold files for loading and unloading: the user stage, table stages, and named internal stages, and external stages over cloud storage such as Amazon S3, Azure Blob Storage, and Google Cloud Storage. Recognise file format objects and the supported formats (CSV, JSON, Parquet, Avro, ORC, XML) and how a storage integration secures external access.

user stagetable stagenamed internal stageexternal stagefile format objectstorage integration

Practice question for this objective

Free sampleData Loading, Unloading, and Connectivitymedium

A security team forbids storing cloud provider access keys in any SQL object and requires that an external stage reach an Amazon S3 bucket using a managed identity rather than embedded credentials. Which Snowflake object lets the external stage authenticate to the bucket this way?

  • AA file format object, because the file format holds the cloud identity used to reach the bucket so the external stage can authenticate to S3 without keys in the stage definition
  • BA network policy, because attaching the bucket to a network policy grants the external stage an identity-based path to S3 that avoids placing access keys in the stage
  • CA secure view over the bucket, because defining the external stage on top of a secure view lets Snowflake reach S3 through a managed role instead of embedded credentials
  • DA storage integration, because it references a cloud IAM role so the external stage authenticates to S3 through that managed identity with no access keys stored in the stage Correct
Use a storage integration to let an external stage authenticate to cloud storage through a managed IAM role rather than embedded access keys. A storage integration is a Snowflake object that references a cloud provider IAM role and the allowed storage locations, letting an external stage authenticate to a bucket through that managed identity; because no access keys are stored in the stage or any SQL object, it satisfies a requirement to avoid embedded credentials.

Why A is wrong: A file format only describes how to parse files and carries no cloud credentials or identity, so it cannot authenticate the stage to S3 as claimed.

Why B is wrong: A network policy restricts inbound IP ranges to the account and has no role in cloud storage authentication, so it cannot provide the identity-based access described.

Why C is wrong: A secure view controls access to query results inside Snowflake and cannot be placed over a bucket, so it offers no path to authenticate an external stage.

Why D is correct: A storage integration delegates to a cloud IAM role and keeps credentials out of the stage definition, which is precisely the keyless managed-identity access the security team requires.

See more COF-C03 practice questions, answers explained.

More in this domain

Back to all Data Loading, Unloading, and Connectivity objectives, or the COF-C03 cert hub.

Examworthy is not affiliated with or endorsed by Snowflake. Original, blueprint-aligned practice material only.