Skip to main content

Build with (drivers, SDKs, programmatic APIs)

Connect from (ingest / stream / ELT)

  • Snowpipe Streaming — programmatic, low-latency row ingest.
  • Kafka Connector — Kafka Connect sink (Confluent/OSS) to Snowflake; supports Snowpipe/Streaming.
  • Spark Connector — read/write Spark DataFrames with pushdown to Snowflake.

Orchestration & DevOps

  • Airflow Provider — hooks/operators/sensors; default conn snowflake_default.
  • Terraform Provider — IaC to manage warehouses, roles, grants, etc.; officially supported v2+.
  • dbt on Snowflake — dbt projects managed via Snowflake CLI & “dbt Projects on Snowflake.”

BI & Analytics

  • Power BI — native connector (GA); supports Microsoft Entra SSO and ADBC-based v2.0 impl.
  • Tableau — native Snowflake connector; OAuth/PrivateLink options; proxy/warehouse guidance.

“Which should I use?” (quick decision flow)

  1. Running SQL in an app/service → pick a language driver (Python/Node/.NET/Go/JDBC/ODBC). If you can’t install drivers or want a polyglot HTTP path, use the SQL API (REST).
  1. Need sub-second streaming ingest → if you already have Kafka Connect, use the Kafka Connector; otherwise, go direct with Snowpipe Streaming.
  1. On Spark/DatabricksSpark Connector for pushdown; consider Snowpark/dbt on Snowflake to move transforms into Snowflake.
  1. Orchestrate & manage infraAirflow for pipelines; Terraform for Snowflake objects; dbt Projects on Snowflake for governed transforms.
  1. BI consumptionPower BI or Tableau native connectors with OAuth/SSO as needed.