Merlin Moncure <mmoncure@gmail.com> writes:
> The downside of SQL-MED, particularly the way postgres implemented the
> driver API, is that each driver is responsible for for all
> optimization efforts and I think this is bad.
There was never any intention that that would be the final state of
things. All the FDW APIs are quite experimental at this point, and
subject to change, and one of the reasons for change is going to be
to improve the optimization situation.
At the same time, it's hard to say what might constitute optimization
for FDWs that aren't backed by a remote SQL database. There are always
going to be reasons why an FDW will have to do some of that work for
itself.
regards, tom lane