SnowFlake - Dhana - Easy

What type of architecture does Snowflake primarily employ?

• Shared-Disk
• Shared-Nothing
• Hybrid Shared-Disk/Shared-Nothing
• Clustered Index

Answer explanation

Snowflake uniquely combines aspects of both shared-disk (centralized data storage) and shared-nothing (independent compute clusters) architectures, allowing for decoupled storage and compute.

Which of the following is NOT a core component of Snowflake's architecture?

• Database Storage
• Query Processing
• Cloud Services
• On-Premise Data Centers

Answer explanation

Snowflake is a fully cloud-native data warehouse and does not require or utilize on-premise data centers for its core operations.

What is a "Virtual Warehouse" in Snowflake?

• A physical data center location
• A compute cluster used for query processing
• A storage unit for data
• A network gateway

Answer explanation

Virtual Warehouses are the compute resources in Snowflake that execute queries and perform DML operations.

Which of these is the smallest size for a Snowflake Virtual Warehouse?

• Large
• Medium
• Small
• X-Small

Answer explanation

The smallest available warehouse size in Snowflake is X-Small, which scales up to 2 servers.

How is data primarily organized in Snowflake storage?

• Row-oriented tables
• Column-oriented tables (micro-partitions)
• Key-value pairs
• Document-oriented collections

Answer explanation

Snowflake stores data in a columnar format within immutable micro-partitions, which are automatically optimized for performance.

Which cloud providers does Snowflake support?

• AWS only
• Azure only
• GCP only
• AWS, Azure, and GCP

Answer explanation

Snowflake is a multi-cloud platform and can be deployed on Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform (GCP).

What is the purpose of Snowflake's "Cloud Services Layer"?

• To store all user data
• To manage virtual warehouses
• To handle authentication, metadata, and query optimization
• To process all user queries

Answer explanation

The Cloud Services Layer acts as the brain of Snowflake, coordinating various operations including security, metadata management, query compilation, and transaction management.

Can a single virtual warehouse process queries from multiple users concurrently?

• No, each user needs their own warehouse.
• Yes, if configured for multi-cluster warehouses.
• Yes, a single warehouse can handle multiple concurrent queries.
• Only for read operations, not write.

Answer explanation

A single virtual warehouse can process multiple concurrent queries from different users. Multi-cluster warehouses provide additional concurrency and elasticity by spinning up multiple clusters of compute resources.

What is "Snowflake Credit" primarily used for?

• Purchasing storage capacity
• Paying for data transfer
• Consuming compute resources (virtual warehouses)
• Licensing user accounts

Answer explanation

Snowflake Credits are the unit of billing for compute resources consumed by virtual warehouses. Storage and data transfer are billed separately.

What happens if a virtual warehouse runs out of compute resources for a query?

• The query will fail immediately.
• Snowflake will automatically scale up the warehouse (if auto-scale is enabled).
• The query will queue until resources become available.
• The query will be re-routed to a different warehouse.

Answer explanation

If a single-cluster warehouse runs out of resources, queries will be queued. For multi-cluster warehouses, new clusters can spin up to handle the load, preventing queuing.