> Suppose I have a table T that has, among its columns, the fields X and Y,
> where Y is an integer, and multiple rows with the same value of X are
> possible. I want to select the rows corresponding to the greatest values of
> Y for each value of X.
You could use DISTINCT ON --- see the "weather reports" example in the
SELECT reference page. It's not standard SQL but will usually be faster
than solutions that use only standard features.
regards, tom lane