I managed to get a plan I was hoping for, but it still doesn't prune
partitions. I created a new operator #|<(integer[], integer) that is
defined in SQL and is basically equivalent to value=ANY(array), and a
non-stable tenants() function defined that returns an array from the
setting, and with that
Good idea. Thanks.I did check. It's not enabled by default but just in
case I did another build. This time explicitly defining --disable-debug and
--disable-cassert. And I tested. Still slower than old versions.
This feels like a build configuration problem. Just can't put my finger on
it yet.
On 10.09.24 15:58, David Mullineux wrote:
I'm getting a bit concerned by the slow performance of generating uidds
on latest dev code versus older versions. Here I compare the time to
generate 50k random uuids. Both son the same machine.
I must be missing something.
Are you sure that the 18dev
most of query sessions using jdbc connections, the one who use ODBC
one savepoint/per statement, but it does not run any "select for update;
savepoint;update", since row lock conflict, so not easy to touch same row
with update/delete, no idea how that create multixact? a MultiXact may
contai
