please
I still do not know what kind of form that a proposed query planner would take. In several RDBMS they:
- silently work all the time on every query to optimize execution patterns based on index size/contents
- power tooling to print detailed execution plans for human consumption (eg:
EXPLAIN) - must be explicitly bypassed (eg
STRAIGHT JOIN)
Can we use this post as a prompt to start a RFC? Or, has one been compiled internal to dgraph that we can peek at?