This is what I mean by updates due to C-Store warehouse engines (Redshift, BigQuery, Snowflake, etc). It's not just that the cloud providers are happy to rent

This is what I mean by updates due to C-Store warehouse engines (Redshift, BigQuery, Snowflake, etc). It's not just that the cloud providers are happy to rent you more space - it's also that they're C-Store and because of that redundancy in columns is well compressed automatically. We try to explain some of that here - https://dataschool.com/data-modeling-101/row-vs-column-oriented-databases/ https://dataschool.com/data-modeling-101/row-vs-column-orien... And Fivetran did a great benchmarking of it here - https://fivetran.com/blog/obt-star-schema https://fivetran.com/blog/obt-star-schema The architecture of the C-Store warehouses often removes the benefits of materialized views. This is why for a very long time Redshift didn't even support them - they insisted they weren't needed as they didn't improve performance over regular redshift significantly.