TupleTableSlot
Also called: tuple slot, result slot, scan slot
A TupleTableSlot is the small holder that carries the current row between executor nodes. It can point straight at a tuple still sitting on a disk page, or hold a freshly built row in memory, without forcing a copy each time a node hands a row upward.
What this means
The small holder that carries the current row from one executor step to the next. The clever part: it can just point at a row still sitting on its disk page rather than copying it, so handing rows up the plan doesn't mean copying data at every step.
Why it matters operationally
Slots are why the one-tuple-at-a-time model stays cheap: passing a row up the tree is passing a slot, not copying bytes. They also defer the expensive step of splitting a stored tuple into individual columns until a node actually needs a column, a key performance trick.