DSM-G600, DNS-3xx and NSA-220 Hack Forum

Soo.. You don't recommend using RAID 1?
I should just use two 750s and backup the important files from one to the other?

if you decide to backup from one to the other there are various ways to automate this which can be found by searching this forum or looking at the wiki

In theory, but I would have a backup of the data just in case.  I personally set up my drives manually using a telnet session; D-Link web GUI has had too many issues formatting the wrong drive for my tastes.

Well.. I don't want to have to purchase another external Drive to back it up to.
I already have a Maxtor 300 GB Network drive that is on it's last legs, so I'm transferring that (TV Shows) to my DNS-323.
Then If I get a 1TB Drive, I can back up my current collection of about 600GB to the 750 drive.. and then anything I don't care about I can leave un-backed up in the excess 250 GB that can't be backed up.

Unless there is a cheaper or smarter method then this? Down time is never an issue.

First thing you need to do is to "de-link" the idea that loss of data is the result of a disk failure - once you can do that, everything else will fall into place.

Disk failure is one of the causes of data loss, but not the only cause - as you point out RAID1 will protect you from data loss cause by a single disk failure - other causes of data loss are users, viruses, and hardware failure - RAID1 offers no protection from any of these.

Now - let's look specifically at the other point you raise ...

Without RAID1, if a disk really fails then I lose all my data since my last backup (assuming I even have a recent one) - no matter how much time I have

Let us assume that the reverse is true - you have no RAID1, but - not only do you have a backup, it is up to date - to the very minute and last transaction before the drive failed, let us also assume that time is of the essence, and you're loosing $50,000/minute - how long will it take you to replace the failed disk, restore the backup and be ready to resume processing?  What is the cost of this down time?

With RAID1 - the processing continues uninterrupted - in most cases the end user never knows a drive failure occurred, only the system administrator - and he/she then takes what ever steps are required to replace the failed drive at a convenient time and with the minimum possible disruption to productivity.

THIS is the purpose of RAID technology - every version of RAID other than RAID0 - THIS is the reason that just about every server sold has a RAID array.

Bigger businesses take this concept further - redundant disks only protect from disk failure - what happens if the server itself fails (memory, motherboard, whatever), you can have redundant servers - two (or more) servers and if one fails the other(s) continue to function and again the end user is completely unaware

Welcome to the world of high availability - where - as much as possible, everything is duplicated, data, disks, memory, power supplies, network cards, the entire server, UPSs, networks, right down to the data center itself.

I have one client that maintains two sites - each equipped with a Windows 2003 server cluster, each equipped with two servers and a total of five (5) RAID arrays between them and tape back up that runs at 15 min intervals, each site has it's own UPS and standby diesel.

Yes.. so this answers my question perfectly..
Let me get this straight:

I should purchase:
a) Another hard drive for my DNS-323
b) Another method to backup my movies
c) A diesel generator to backup my movie collection.

Thanks!