Analytics Software

Hi all,
.

.
SQL> select * from TEST6152;

         A B   DD
---------- --- ---------
         2 USD 12-DEC-07
        30 USD 12-DEC-07
        30 USD 12-NOV-07
        15 USD 22-NOV-07
        65 USD 13-SEP-07
.
.
.

I require this output from the quiery
.
CUR DT          SUM(T.A)
--- --------- ----------
USD 30-SEP-07         65
USD 30-NOV-07        110
USD 31-DEC-07        142
.
.
which is basically the cumulative sum
.
Now this is what I tried
.
.
SQL> ed
Wrote file afiedt.buf

  1  select t1.curr, t1.dt, sum(t.a) from
  2  test6152 t,
  3  (
  4  select B curr, last_day(trunc(DD)) dt from TEST6152 group by B, last_day(trunc(dd))
  5  ) t1
  6  where
  7  t.b = t1.curr
  8  and t.dd < t1.dt
  9  group by
 10     t1.curr, t1.dt
 11* order by 1, 2
SQL> /
.

CUR DT          SUM(T.A)
--- --------- ----------
USD 30-SEP-07         65
USD 30-NOV-07        110
USD 31-DEC-07        142
.
.

Now my question is : Is there a better way to do this in SQL coz when I translate the same logic to a bigger table (rows > 3 million) , it gives me a performance issue.

Thanks in advance