I worked on a MySQL-on-RDS setup once where the company was splurging (and being charged through the nose, ofc) on a backing EBS volume with 20k IOPS. Meanwhile nobody had bothered to update the default settings, which limited all background work to use no more than 100 IOPS. Needless to say, this DB had severe performance problems.
Managed databases in the cloud are a bit like ORMs: they significantly reduce the amount of knowledge needed 99% of the time. Sadly, when the 1% of the time hits this also means that nobody on the team has built up the DBA skills required for the big hairy problem. I see this pattern repeated all the time when consulting for startups that just had their first real traction and now the magic database box is no longer working as before.
Managed databases in the cloud are a bit like ORMs: they significantly reduce the amount of knowledge needed 99% of the time. Sadly, when the 1% of the time hits this also means that nobody on the team has built up the DBA skills required for the big hairy problem. I see this pattern repeated all the time when consulting for startups that just had their first real traction and now the magic database box is no longer working as before.