Large File Revision Control??

[BCC]

Hi,

We are currently using cvs for some code versioning control, and would
like to expand it a bit but are running into problems...

What we have are many MySQL tables that must be synchronized across a
network ( this is on linux btw ).  i.e. There is one development
machine, and when I or whoever is working on the DB would like, we can
update 2 other machines with the same databases. Replication was our
first attempt, but since the master was also being used for development,
replication kept crashing.  

Next I thought that it would be great to use cvs- just add my database
into the repository and then when it is clean on the 'master' I would
simply go update it on the other machines.  

But NO!  I couldn't do it, our files are too large.

Several of our database tables are ~800 megs.  Since the machine I am
using to try and add the repository only has 512 megs of ram, I cannot
add the repository.  I get various error messages, from out of memory
(if I try to add many files), to no space left on device when adding one
249 meg file (I think this is because the swap gets used up?).

How can I do this?  The machines that I will be checking out the
repository from each have 2 gigs of ram, so once I have these big files
in the repository it should be okay (?).  I can have the 512 machine
upgraded to 2gigs as well. 

Are there better ways to do this in linux?  Short of tar, gzip, and scp?

Thanks,
Bryan