> Index the //independent// `where` predicates to improve hash join performance.

> Indexing join predicates doesn’t improve hash join performance.

> Indexing a hash join is independent of the join order.

> Select fewer columns to improve hash join performance.

> Hash joins do not need indexes on the join predicates. They use the hash table instead.
> A hash join uses indexes only if the index supports the independent predicates.
> Reduce the hash table size to improve performance; either horizontally (less rows) or vertically (less columns).
> Hash joins cannot perform joins that have range conditions in the join predicates.