In a separate post, _dave_ said:
"Heavy" users should not install Postgres to a seperate drive (or array) - they should install Postgres to "Program Files" like any other user, and initialise the cluster on their other drives. That or ideally (unless the "Program Files" drive is shockingly slow) use tablespace to store *data only* on the fast / large disk(s).
I have a similar setup to the OP of this thread, with a smaller older internal drive and a new, massive 1 TB external drive (connected via USB). I also have 2 other networked computers that will access this data. Some questions:
1) Why should we install PostGres to "Program files" or the main "c:" drive and the *data only* to the fast / large disk? Obviously I want to install the data on the larger external drive, but what is the advantage of putting the PostGres (application?) on the main drive? Given the external drive is connected via USB, I imagine the access time is faster to the internal drive.
2) Also, how portable would this configuration be? Could I unplug the external drive, and plug it into my laptop and access all of the same data?
thanks,
fooz