Skip to content

sfc-gh-tspann/Ingest

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 

Repository files navigation

Snowflake connection methods (REST, Python, JDBC, ODBC) and Uses

General Best Practices https://www.snowflake.com/en/blog/best-practices-for-data-ingestion/

Snowpipe Streaming https://www.snowflake.com/en/blog/data-ingestion-best-practices-part-three/

Snowflake Connector for Apache Kafka https://www.snowflake.com/en/blog/best-practices-for-data-ingestion-part-2/

All data going over HTTPS all more or less perform at similar rate limits.

Single Thread -> Single Connection -> Single Snowflake account -> per warehouse-> ~60 queries per second

Fastest Way to Send Data to Snowflake

  • PUT INTO a Stage
  • COPY INTO a Table

https://www.snowflake.com/en/blog/best-practices-for-data-ingestion/

DevOps / Administration / Automation

  • REST API - for DDL, administration, monitoring, management

https://docs.snowflake.com/en/developer-guide/snowflake-rest-api/snowflake-rest-api

  • SQL REST API - for adminstration, kick off stored procedures, do not use for ingest.

https://docs.snowflake.com/en/developer-guide/sql-api/index

Good use for REST API is to trigger Snowpipe to read files from stages.

image

Check Your Network Speed

https://docs.snowflake.com/en/user-guide/snowcd

Drivers

Drivers will differ based on the language, but the majority of speed is the same as HTTPS is the same.

https://docs.snowflake.com/en/developer-guide/driver-connections

Download your tools

https://docs.snowflake.com/en/user-guide/snowflake-client-repository

Libraries -- Snowpark

Libraries -- Snowflake Native App

Libraries -- External Functions

Libraries -- Apache Spark, Apache Flink, Apache NiFi

Suggestions

For fast data ingestion into Snowflake, using Snowpipe with gzipped CSV files (or other compressed formats) is generally considered the fastest and most efficient method, especially for batch loading.

Releases

No releases published

Packages

No packages published