We are having good luck with a historical look for disk usage. If you average the growth over the past 90 days, you can extrapolate that as far into the future as you want. That factors in growth spikes as well since it will increase the growth rate. We calculate a growth per day so when you divide current free by growth per day, you get the estimated number of days before things blow up. That has worked better for us than the typical 10, 20, whatever% free since we have some systems that would blow through 20% free in a matter of hours due to special projects. The "days left" report as we call it allows us to perform capacity planning twice a year (and thinking about changing to once) and a daily report that alerts us to anything withing 60 days of running out of space. Not perfect, but has kept us out of trouble for close to 1 year with new databases, old databases and new projects and database consolidations. Now if we can write some scripts to track memory, network and CPU consumption, we will be set. On 2/23/06, Tanel Põder <tanel.poder.003@xxxxxxx> wrote: > > > Capacity planning can be divided into disk, cpu and memory planning. > > .. and also contention planning.... > > Tanel. > > -- > //www.freelists.org/webpage/oracle-l > > > -- Stephen Andert http://spaces.msn.com/members/andert-news