Question

I have a SQL db with several million triples stored. A large portion of the triples are repeated at least once. What is the most practical way to count the number of times that every triple occurs in the db? Thanks.

Answer 1

Supposing that the “triple” is made up of three columns c1, c2, and c3, you can use group by to count the distinct tuples:

  select c1, c2, c3, count(1)
    from mytable
group by c1, c2, c3

Demo: http://www.sqlfiddle.com/#!7/693c4/2