I have a new 150GB hd that I would like to configure as the master and a my old 10GB hd as the slave. What kind of partition scheme should I use? I would like to use the Dump and Restore package to create a viable back up option to my 10G HD. What specifically should I back up in order to be able to recover from a hard drive failure or other issue? If I have development source code stored in PVCS and DB2 what would be vital to back up for these PKG's. Does the Dump pkg actually compress the data before it is stored as backup.

Partition scheme : It depends on the use of the computer.
I.e: One of my servers (Oracle 9 database)

/ 4 GB
swap 2GB
/opt 8 GB
/home 3 GB
/u01 6GB
/u02 6GB
/u03 6GB
/u04 6GB

Then sum up all the extra space to /opt (and/or /opt2), or leave it for future use.

You can try LVM. It lets you grow partitions "on-demand" from unpartitioned disk. With newer distros is pretty easy to setup.

-Dump uses compression if you ask to (-j switch) (dump -j9 ...)

- If you only have 10GB for restoring 140GB, forget about recovering the whole disk "as is". Just save your valuable data (/etc, /root, /usr, / --this last one specified by filesystem (i.e.: /dev/hda1) or else you will copy whole disk as the whole disk is mounted under / )
If still you have place for more, include /opt or whatever).

Thank you for this advice.
I will implement according to your recommendations. I am also condering backing up this server to a shared network drive.

If I choose to map a drive to our network and store the backup would it make any difference if it was a Windows box or a Unix box?

the backup can be saved to a device (i.e. a DAT) or to a file (i.e. bk1.dmp)

If you choose 2nd way, you can store the resulting file whever you want, just taking care not to break it (i.e. ftp'ing on ascii mode will break it). Just copying via samba will be ok.