Quick & Dirty: What is a Snowflake Database?

Over the last few years the demand for a new database built in the cloud has skyrocketed. In 2020 Snowflake released their IPO and the word has only spread further and faster with practically only high praises. With other cloud databases in the market, why would a new one that comes in relying on cloud platforms such as AWS, Azure, and GCP take off so quickly? The short answer, because it’s better.

Snowflake is not an ordinary database by any means and has managed to massively disrupt the traditional data warehouse mindset. Although many of the same principles apply, a lot of the work and maintenance has been off-loaded into the cloud. Yes, Amazon has Redshift and Azure has Azure Data Warehouse, but Snowflake is still arguably a better cloud database. Plus, it requires little to no maintenance once it’s up and running. Users can now plugin their analytics tools such as Tableau or Power BI and know that Snowflake will be able to perform.

Forget the traditional practice of indexing tables for improved query performance. Snowflake does that for you! Yes, we can still do traditional dimensional modeling, but it’s no longer a necessary requirement to start doing analytics. Snowflake isn’t your traditional relational database either. Snowflake is a columnar database that has a modified version of SQL that allows us to query against semi-structured datasets such as JSON, Avro, ORC, and Parquet.

Final Thoughts

Overall Snowflake has made an impressive introduction and the other cloud providers have some catching up to do. The pricing model is very flexible and automatically shuts off services when they aren’t being used. Compute tiers can be changed at a click of a button making the need for a Snowflake DBA practically non-existent. It’s no surprise so many companies have jumped ship to make Snowflake their go to database solution.