> The scaling out solution pretty much these days for SQL database is replication, sharding and partitioning tables. Postgres does these very well. I tend to d

> The scaling out solution pretty much these days for SQL database is replication, sharding and partitioning tables. Postgres does these very well. I tend to disagree that Postgres does partitioning well; partioning table resolution is rather dumb, there is no hinting (e.g. there is no way to tell postgres that there is a unique mapping for each query to a table) and no caching of CHECK results.