Snowflake

Definition

Snowflake is a cloud-native data warehouse platform that separates storage and compute so organizations can scale each independently, running SQL analytics on massive datasets across AWS, Azure, and Google Cloud.

Overview

Snowflake was founded in 2012 by data-warehousing veterans and launched publicly a couple of years later, among the first data warehouses architected specifically for the cloud rather than adapted from on-premises database engines. It went public in a high-profile 2020 IPO. Its multi-cluster shared-data architecture stores data once in cloud object storage, such as Amazon S3, while allowing multiple independent compute clusters — "virtual warehouses" — to query that same data simultaneously without contention. Compute can be scaled up, down, or paused entirely, independent of storage costs. Snowflake supports semi-structured data like JSON and Parquet alongside traditional relational tables, and has expanded into secure cross-organization data sharing, Snowpark for running Python, Java, and Scala code against warehouse data, and Cortex for built-in AI and LLM functions. Snowflake is widely adopted for enterprise analytics and data lake consolidation, often paired with transformation tools like dbt and orchestration tools like Apache Airflow in a modern data stack, and it competes directly with BigQuery and Amazon Redshift.

Key Features

• Multi-cluster shared-data architecture separating storage and compute
• Independently scalable "virtual warehouses" that can pause when idle to save cost
• Native support for semi-structured data such as JSON and Parquet alongside SQL tables
• Secure data sharing and a marketplace for cross-organization data
• Snowpark for running Python, Java, and Scala code directly against warehouse data
• Cortex AI functions for built-in LLM and machine-learning capabilities

Use Cases

Frequently Asked Questions

Related in Database