Re: Hardware HD choice...

If you are doing batch inserts of data, and want to have reporting queries concurrently running, make sure you have the pg_xlogs on a different disk than the data/indexes.   2 drives RAID 1 for OS + xlogs works great (and these can be SAS if you choose, have a separate partition -- ext2 if it is linux -- for the xlogs.  Then you can easily go with storage capacity and SATA for the main reporting portion.  You just don't want the inserts in batches to slow the whole thing to a crawl due to xlog writes on the same drive array as the reporting.
However, if you can only get a few disks, it is a lot harder to choose between one large array and two of them split without experimenting with both on real data and queries.  It is a quick and easy performance win if you have 6+ disks and do enough writes.

Also, if you intend to have lots of data organized by a time field, and expect to do the reporting/aggregation queries on subsets of that data bounded by time, partitioning by time can have huge benefits.  Partition by month, for example, and sequential scans will only flow to the months of interest if the queries have the right lmits on the date in the where clause.

Partitioning WILL take more development and tuning time, so don't do it unless you know you need it... though if the reporting is mostly restricted to time windows, the impact it has on improving runtimes of aggregation queries is immense.  However, partitioning won't help at all until you have enough data to justify it.