The join across machines
students is sharded by student_id, and colleges is a small table
of 400 rows. A query joins them.
What is the cheapest arrangement?
students is sharded by student_id, and colleges is a small table
of 400 rows. A query joins them.
What is the cheapest arrangement?