What if I have table X
(heap table) that is being frequently updated (INSERT
and DELETE
only), and table Y
which can never exceed 2000 rows. Table X
has a foreign key column that references a unique index of table Y
, and the relationship type is one-to-many. However there can only be maximum of 10 foreign keys per unique index, which means that table X
can never exceed 20000 rows. Basically table X
can have any amount of rows between 0-20000 at any moment. Makes sense?
So is heap table good there or should I add a clustered index there? Or maybe someone has an even better solution?