Going through you request, i found that Table Partitioning might do the trick for you.. Me, in my sytem, implemented this Partition for Archive data; which is maintaining only the latest set (we have specific mechanism on we define sets using "versions" each time a job inserts data into the Archive tables) of data for each month in one filegroup and then "moving" the rest to the some another filegroup which will be located on a different drive..

If you are interested in this process, try going this awesome whitepaper from Kim Tripp : Partitioned Tables and Indexes in SQL Server 2005

Hope this helps! 🙂