Selecting random rows efficiently

Hi,
i have a table of around 3 million rows from which i regularly (twice a second
at the moment) need to select a random row from

currently i'm doing "order by rand() limit 1" - but i suspect this is
responsible for the large load on my db server - i guess that PG is doing far
too much work just to pick one row.

one way i can think of is to read in all the primary keys from my table, and
select one of the keys at random then directly fetch that row.

are there any other ways to do this? i need to keep the load down :)

Thanks,
Richard