1. Depends on your use case. There's
1. Depends on your use case. There's overlap in the problems they solve, certainly, but there's also reasons you might want both.
2. Yes, the issue with caching returning empty results was resolved quite a while back
3 Replies
One thing regarding replicas is that Hyperdrive may not route to your replicas as you wish it did globally since the connection pooling is not currently in many regions. Some folks are managing Hyperdrive connections to their replicas in order to proper route requests to replicas, and this is something we're aware of and thinking about
Ah interesting, thank you. Honestly I didn't even realise it was possible to control routing to replica vs. master DB - I thought that was taken care of by the DB platform provider - in my case Azure. As they tell it, read queries go to replicas and write queries go to master, without me doing anything, but that's probably showing some ignorance on my part.
I'll try it out with Azure, they may handle things differently, the manual routing was a solution that someone else had found for AWS RDS postgres