DevOps Articles

Top 3 Snowflake Performance Tuning Tactics

by The Snowflake Data Warehouse has an excellent reputation as an analytics platform for blisteringly fast query performance but without indexes.

Like many other data analytics platforms, Snowflake uses a columnar data store which is optimized to fetch only the attributes needed for the specific query, and this is illustrated in the diagram below: In the above diagram, the query fetches just two columns, and on a table with 100 columns, this will be 98% faster than a traditional row-store, which needs to read all the data from disk. The diagram below illustrates a vital component of the Snowflake internal architecture that it caches data in both the virtual warehouse and the cloud services layer.

By all means, consider scaling up to a large warehouse to improve query performance, but first identify and focus on the actual problem.