News for hackers. Stuff that matters.
Re:SparkFun's migration from MariaDB to PostgreSQL by vidarh on Wednesday November 13, 2013 @06:23AM (#6724495)
> * Postgres is generally considered to scale to many concurrent connections and many cores. Useful for OLTP.

But MySQL scales trivially to many servers. Postgres replication is finally getting there, but it's been incredibly slow coming, while setting up huge MySQL installations with a dozen+ replicas, with selective sharding etc. has been effortless for about a decade.

Re:SparkFun's migration from MariaDB to PostgreSQL by damncabbage on Wednesday November 13, 2013 @09:08AM (#6725194)

  ... while setting up huge MySQL installations with a
  dozen+ replicas ...
If you're careful about your function calls, sure. Include a UUID() or SYSDATE() in your INSERT statements, and the replication goes to pieces as the function calls are run separately on each slave (with differing results).
Re:SparkFun's migration from MariaDB to PostgreSQL by morgo on Wednesday November 13, 2013 @09:18AM (#6725270)
I recommend switching to Row-based Replication, which avoids this problem.