Understand why a B-tree index accelerates equality and range lookups compared to a full table scan.
You run SELECT * FROM users WHERE email = 'a@b.com' on a large table with a B-tree index on email. Why is this typically much faster than the same query without the index?