SnowPro Associate: Platform Question 6
Single answer● Elastic storageA retail company stores three years of sales history in Snowflake. Most analysts query only the most recent 90 days, but finance runs month-end reports that sometimes access older data. The data volume is growing rapidly, and the team wants to reduce operational effort while ensuring that storage can grow without planning disk capacity in advance. Which Snowflake capability best addresses this requirement?
- A
Snowflake automatically separates compute from centralized storage, allowing storage to scale independently without provisioning additional warehouse disk capacity.
- B
Snowflake requires each virtual warehouse to reserve local disk space for the tables it queries, so increasing warehouse size is the main way to scale storage.
- C
Snowflake can store only active table data; historical versions for Time Travel and Fail-safe must be exported to external cloud storage to support growth.
- D
Snowflake storage scales only when multi-cluster warehouses are enabled, because additional clusters provide the persistent capacity needed for older data.
Show answer and explanation
Correct answer: A
Explanation
The key concept is Snowflake's separation of storage and compute. Data is stored in centralized cloud storage, while virtual warehouses provide independent compute for loading and querying. Because of this architecture, storage is elastic and can grow without administrators planning disk capacity on specific compute nodes. This is especially useful in scenarios where data volumes increase over time but query patterns vary, such as frequent access to recent data and occasional access to older history. A common misconception is that warehouse size or multi-cluster configuration affects storage capacity; in Snowflake, those settings affect compute performance and concurrency, not persistent data storage. Snowflake documentation on architecture and virtual warehouses emphasizes that storage and compute scale independently, which is a foundational platform concept for the SnowPro Associate exam.
- A. Correct.
Correct. Snowflake uses a shared data architecture with centralized cloud storage that is separate from compute resources. This elastic storage model means data volume can grow without requiring customers to provision or manage storage attached to virtual warehouses. Warehouses provide compute for query execution, not persistent table storage.
- B. Incorrect.
Incorrect. This reflects a traditional on-premises or tightly coupled MPP misconception. In Snowflake, virtual warehouses do not own persistent table storage and do not need local disk reservations for the data they query. Resizing a warehouse increases compute resources, not long-term storage capacity.
- C. Incorrect.
Incorrect. Snowflake manages table data along with historical data used for features such as Time Travel and Fail-safe within its storage layer. Customers do not need to export older table versions to external storage as part of normal elastic storage management. This option confuses Snowflake-managed storage with optional external stages or archival patterns.
- D. Incorrect.
Incorrect. Multi-cluster warehouses improve concurrent query processing by adding compute clusters, but they do not provide persistent table storage capacity. Storage elasticity in Snowflake is independent of whether a warehouse is single-cluster or multi-cluster.