Postgres Just Cracked the Top Fastest Databases for Analytics

Re:Postgres Just Cracked the Top Fastest Databases for Analy... by bhouston on Saturday March 08, 2025 @07:27AM (#43299679)

> What do you mean by "standard" SQL?

No extensions, just the default Postgres Cloud SQL offering from Google Cloud.

> Also, what makes BigQuery not a "standard SQL database" -- they use GoogleSQL, true, but many DBs have slight SQL variants.

I view BigQuery as not a standard DB because it really doesn't do transactions, rollbacks, foreign key constraints, indices, stored procedures, etc. I view it as a non-SQL database pretending to be SQL for query convenience.

[ Parent | on HN ]

Re:Postgres Just Cracked the Top Fastest Databases for Analy... by gbin on Saturday March 08, 2025 @07:42AM (#43299784)

I think you mean a relational database.

[ Parent | on HN ]

Re:Postgres Just Cracked the Top Fastest Databases for Analy... by layer8 on Saturday March 08, 2025 @07:58AM (#43299888)

All the features mentioned are part of SQL.

[ Parent | on HN ]

Re:Postgres Just Cracked the Top Fastest Databases for Analy... by nom on Saturday March 08, 2025 @08:21AM (#43300046)

SQL is a language to interact with a relational database. It is not about the language.

[ Parent | on HN ]

Re:Postgres Just Cracked the Top Fastest Databases for Analy... by jandrewrogers on Saturday March 08, 2025 @09:13AM (#43300404)

SQL (the DML) is just a query language. It could be implemented on anything.

Parts of SQL DDL is implementation specific. The SQL standard hardcodes the assumption of (by modern standards) a small-scale system and therefore will deviate significantly on large-scale systems. There is no practical way to make this "standard", you have to understand the implementation it reflects.

[ Parent | on HN ]