Really nice to see the direction things are moving in... I do feel that the replication/failover story needs a lot of work, but there's been some progress towards getting it in the box. Even digging into any kind of HA strategy is cumbersome to say the least (short of a 5-6 figure support contract). It's one of the things that generally stops me from considering PostgreSQL for a lot of projects. As a side note, I really like how RethinkDB's administrative interface is and their failover usage. It would be great to see something similar reach an integration point for PostgreSQL.
I also think that PLv8 should probably make it in the box in the next release or two. With the addition of JSON and Binary JSON data options, having a procedural interface that leverages JS in the box would be a huge win IMHO. Though I know some would be adamantly opposed to this idea.
I also think that PLv8 should probably make it in the box in the next release or two. With the addition of JSON and Binary JSON data options, having a procedural interface that leverages JS in the box would be a huge win IMHO. Though I know some would be adamantly opposed to this idea.