This post was originally written in 2009
In the past I’ve written about hard-disk failures on this and/or other blogs. A hard-disk failure is a pain in the proverbial normally, especially if you forgot to make regular data backups. (As most people, unwisely, do, it seems.) Even if you’re running a RAID configuration using hot-swappable SATA disks a disk-failure causes unwanted hassle and replacement-expense that one could do without.
Either:
1) Have a smaller drive, 250GB or suchlike, fitted in addition to the massive-sized drive, on which resides your operating system and its immediately-related files: ( e.g. drive C: . The huge drive could be drive D: for example. ) . Also take regular backups of both drives.
2) Run a RAID configuration in RAID 1 or greater using 2 or more SATAII large disks (>500MB) of equal size. Take regular backups.
OR 3) A combination of both. (Inc. regular backups.)
Reasons for 1):
If your operating system screws up, gets a virus infection, or you just decide to upgrade it; it’ll be much simpler to have all the files and folders concerned separately contained on a single, smaller drive, while the massive drive exists independently of the system drive and is used for storage alone.
Whilst it’s true that there’s nothing whatsoever to prevent an invading piece of malware attacking the massive drive at the same time as the smaller C: drive, the computer is able to operate with a reinstalled operating system on C: while you are cleaning your massive data/music/porn storage disk, or formatting it and reinstalling from a backup made prior to the malware infection; hence less down-time than there would have been if you’d had to reinstall all your massive amounts of data plus your OS on a single drive.
Reasons for 2):
A RAID 1 or greater configuration will automatically provide some form of backup, so that should a drive fail, or data corruption occur on one drive, there’ll always be a second drive installed as backup. This type of configuration will prevent data corruption to a greater extent, and, in the case of a drive failure, will allow you to hot-swap the faulty drive for a new one with minimal if any down-time.
In the case of a virus infection however, this approach can be problematic as the entire contents of a terabyte or more will need to be cleaned or reinstalled from a backup made prior to infection. – Thus causing a problem of down-time.
This brings me on to the last section: -
Reasons for 3):
Firstly, we never properly specified what 3) actually is exactly: We just said that it’s a combination of 1) and 2). Let’s be a little more specific: -
The setup is as follows:
2 or more x 250 GB or similar drives in a RAID 1 or greater configuration containing the OS/C: drive, and associated files.
2 or more x 500GB+ drives in a RAID 1 or greater configuration for storage of your other files.
Most motherboards only have a single RAID controller built into them. The above involves 2 RAID configurations, each on a separate RAID controller. Fortunately the remedy; can be found in installing a separate RAID controller. It isn’t that difficult or expensive.
A VIA VT6421 card is the cheapest option, and fits into a spare PCI slot. Silicon Image also have a RAID controller card that fits into that little-used PCIe x1 slot. Both have their own on-board BIOS although they both individually require an extra driver.
In the case of the Silicon Image card you’ll find this on the installation CD that is packed with it. It has drivers for pretty much every known OS included on it, including a multitude of flavours of Linux.The VIA VT6421 driver can probably be installed with or from Windows XP; but if not it also comes with a driver CD too.
So having got our RAID controller card fitted we start off (– Assuming that this is a new build.) by fitting 2 or more identical 250GB drives, connecting them to the motherboard’s RAID controller, and configuring then in RAID 1 or greater, depending upon the number of drives, in tandem with installing the operating system on this configuration.#
Having got the machine working we now fit 2 or more identical 500GB+ drives and, ensuring that the RAID controller card’s drivers are installed and working properly, we connect the big drives’ SATA connector leads to the RAID controller’s SATA ports and reboot…
Now you’ll see 2 drives in My Computer: Drive C: , which is the set of 250GB system drives in RAID configuration containing the operating system and associated files, and Drive *: , which is the set of big drives in RAID configuration.
The drawback of this method is cost: Let’s assume that you configure 2 x 250GB drives in RAID1 , as well as 2 x 1TB drives in RAID 1. That’s going to cost you about £90 extra, at time of writing, than just having a single 1 TB HDD installed. (Including the cost of the extra RAID controller card.)
Question: Is it worth it?
For the peace of mind, and the reduced hassle when it does screw up, I’d say it probably does balance out with the extra cost, and is worth doing. – Particularly if you have loads of data that is going to cause problems if it can’t be accessed instantly.
Do I use this system? I didn’t have enough data on disk at time of writing to justify it. Currently, though, my CD and DVD collection (Music and data.) has pretty much all been put on disk to avoid it being binned along with the failing optical disks. – So yes I imagined at the time I will eventually use this system – but I still don’t: I use an online backup instead.
Should you use this system? Only you can decide that. Please comment and tell us why you’d use it or why you wouldn’t.
You’re looking at Kkomp.com – Beyond. <--Link to Home page.
The URL of what you see is http://kkomp.com/2011/08/22/preventing-data-loss-associated-with-big-disks/
Why not share the linked URL by pasting it into Facebook or sending it by email?
Post Footer automatically generated by wp-posturl plugin for wordpress.















