Hi developers!

I have some questions I guess not answered yet...

As is obvious, our applications begins very small, and when they
grow, anytime we need to do some SQL queries that 'merge' two or
more .gdb's.
Is possible to do it with FB? (This can increase the total data that
can be managed on a system, besides file limits of FB or file system,
and is easier to control / backup / distribute too...)

Example:
One system have academic data, other system controls physical space
over the Institution, on two gdbs.
Can I use the 'RoomID' from Physispace as a foreign key on
Academic ?? Can I answer: What teachers will work at Building X?

Select Academic:teacher.Name from academic:teacher
where academic:teacher.RoomID in
( select Physispace.Rooms.RoomID from Physispace:Rooms where
physispace:Rooms.BuildID = X )

ACADEMIC.GDB -------------------------
----- TEACHERS Table -----
TeacherID integer
Name varchar(50)
RoomID integer
----------------------------------------------------

PHYSISPACE.GDB -------------------------
----- CAMPUS Table -----
CampusID integer
Campus Name varchar(50)

----- BUILDING Table ----
CampusID integer
BuildingID integer
Building Name varchar(50)

----- ROOMS Table -----
CampusID integer
BuildingID integer
RoomID integer
Room Description varchar(50)
-------------------------------------------------------

Can it be done?

Second question, in some articles I saw that the greater databse
report is near 980Gb, is it in only one file? What about performance,
machine type/specifications? It's really hard to make decisions with
not complete data; this kind of truncated info is bad for all of us!
Do you have some names of companies that use FB / IB seriously (heavy
users)?

Third (and last): Is FB ready for paralell multiprocessing? I have a
UNISYS with new CMP architeture, with 08 processors (can be 32!), the
load will be distributed over processors or the FB engine will
have 'affinity' with only one processor?

Txs in advance, and my apologies by my bad english!
Alessandro