File Server Builder's Guide
by Zach Throckmorton on September 4, 2011 3:30 PM ESTHard drives
One of the most frequently asked questions I hear is 'what's the most reliable hard drive?' The answer to this question is straightforward - the one that's backed up frequently. Home file servers can be backed up with a variety of devices, from external hard drives to cloud storage. As a general guideline, RAID enhances performance but it is not a backup solution. Some RAID configurations (such as RAID 1) provide increased reliability, but others (such as RAID 0) actually decrease reliability. A detailed discussion of different kinds of disk arrays is not within the scope of this guide, but the Wikipedia page is a good place to start your research if you're unfamiliar with the technology.
As for hard drive reliability, every hard drive can fail. While some models are more likely to fail than others, there are no authoritative studies that implement controlled conditions and have large sample sizes. Most builders have preferences - but anecdotes do not add up to data. There are many variables that all affect a drive's long-term reliability: shipping conditions, PSU quality, temperature patterns, and of course, specific make and model quality. Unfortunately, as consumers we have little control over shipping and handling conditions until we get a drive in our own hands. We also generally don't have much insight into a specific hard drive model's quality, or even a manufacturer's general quality. However, we can control PSU quality and temperature patterns, and we can use S.M.A.R.T. monitoring tools
One of the most useful studies on hard drive reliability was presented by Pinheiro, Weber, and Barroso at the 2007 USENIX Conference on File and Storage Technologies. Their paper, Failure trends in a large disk drive population, relied on data gleaned from Google. So while the controls are not perfect, the sample size is enormous, and it's about as informative as any research on disk reliability. The PDF is widely available on the web and is definitely worth a read if you've not already seen it and you have the time (it's short at only 12 pages with many graphs and figures). In sum, they found that SMART errors are generally indicative of impending failure - especially scan errors, reallocation counts, offline reallocation counts, and probational counts. The take home message: if one of your drives reports a SMART error, you should probably replace it and send it in for replacement if it's under warranty. If one of your drives reports multiple SMART errors, you should almost certainly replace it as soon as possible.
From Pinheiro, Weber, and Barroso 2007. Of all failed HDDs, more than 60% had reported a SMART error.
Pinheiro, Weber, and Barroso also showed how temperature affects failure rates. They found that drives operating at low temperatures (i.e. less than 75F/24C) actually have the highest (by far) failure rates, even greater than drives operating at 125F/52C. This is likely an irrelevant point to many readers, but for those of us who live further up north and like to keep our homes at less than 70F/21C in the winter, it's an important recognition that colder is not always better for computer hardware. Of use to everyone, the study showed that the pinnacle of reliability occurs around 104F/40C, from about 95F/35C to 113F/45C.
From Pinheiro, Weber, and Barroso 2007. AFR: Annualized Failure Rate - higher is worse!
Given the range of temperatures that hard drives appear to function most reliably at, it might take some experimentation in any given case to get a home file server's hard drives in an ideal layout.
So rather than answering what specific hard drive models are the most reliable, we recommend you do everything you can to prevent catastrophic failure by using quality PSUs, maintaining optimal temperatures, and paying attention to SMART utilities. For such small sample sizes as a home file server necessitates, the most important factor in long-term HDD reliability is probably luck.
Pragmatically, low-rpm 'green' drives are the most cost-effective storage drives. Note that many of the low-rpm drives are not designed to operate in a RAID configuration - be sure to research specific models. The largest drives currently available are 3TB, which can now be found for as little as $110. The second-largest capacity drives at 2TB generally offer the best $/GB ratio, and can regularly be found for $70 (and less when on sale or after rebate). 1TB drives are fine if you don't need much space, and can sometimes be found for as little as $40.
Facebook
Twitter
RSS
152 Comments
View All Comments
HMTK - Monday, September 5, 2011 - link
We don't even use encryption for our enterprise customers. There's just no point. How easy do you think it is recovering useful data from a RAID member? Not talking about a mirror here but RAID 5/6/10qoonik - Sunday, September 4, 2011 - link
Hardware:motherboard: supermicro X7SPA-H-D525 (
- Intel® Atom™ D525 (Passive cooling)
- Intel® ICH9R Express Chipset
- 6x SATA (3.0Gbps)
ram: 2 x 4GB DDR3 SO-DIMM Kingston
case: CFI A7879 Mini-ITX NAS/Server Case - 4 Hot Swap Bays
psu: FSP120-50GNF (FANLESS)
fan: BeQuiet SilentWings 120 mm PWM
hd: 1 x WD green 1.5 TB (completing ... )
OS:
Amahi Home Server (Fedora) http://www.amahi.org
Rick83 - Sunday, September 4, 2011 - link
Gonna chime in here ;)motherboard is a gigabyte P55-UD5
i5 650
10xSATA + 2x eSATA + 1xIDE + 2xIDE from PCI card
2x1 GB some-DDR3
CM Stacker STC-1 (yes, the original goodness!) with 3 4in3 modules
psu: seasonic 430W with plenty of power adapters
fan: stock fans all around (120mm per 4in3, one 120mm exhaust, one 80mm top exhaust, no CPU fan), CPU cooler is a scythe Yasya
hd: 1x 8GB transcend IDE flash module (SLC), 1x 2.5" transcend IDE ssd (SLC), 1x 40GB seagate IDE, 1x 80 gb WD IDE, 1x 80GB seagate SATA, 3x 400GB Seagate SATA, 5x WD 1TB EARS SATA, 1x Samsung 1TB SATA (12 spinning disks, 2 SSDs)
an optical drive
a TV-card (though apparently broken...stream coming out of it is corrupted)
an IEEE1394 CF Card reader
Setup is 2x RAID1 (second level back up, and dynamic system files, consisting of the small disks) ad 2x RAID 5 (main and back-up array)
graphics: nVidia 6200 - looking to replace this with something that idles at lower energy - can't stand the card being as hot as it is, with no screens attached.
Now having lived with that machine for a few years (previously it was running only 7 disks and a sempron single core on it, before that there was a pre-cursor server that was running on different celerons and as little as 32MB of RAM) I am currently looking at options to make the disks more accessible. Something this article doesn't touch on, is with many disks come many deaths. Which is where a hot-plug cage really comes into its own. So I'm on the look-out for affordable backplanes with 120mm fans which I can replace 1:1 with two of the 4-in-3s (my IDE disks will have less use in a hot swap cage ;) )
Death666Angel - Sunday, September 4, 2011 - link
That's a nice coincidence. I just ordered my file server stuff this week and it got here on Friday. So far I just put it together, haven't turned it on, yet (exam stress).I haven't read this article, hope my stuff isn't too useless. ;-)
Just a fyi, here is the system I have:
- Sharkoon T9, 9 x 5.25"!!
- AMD Phenom II X4 840 (fake Phenom btw)
- Asus M4A88TD-V EVO/USB3, which supports ECC memory
- 2x4GB Kingston DDR3-ECC 1333MHz RAM
- 2 x IcyDock 5 in 3 Backplane (MB455SPF-B)
- Intel Gigabit CT Desktop Adapter for PCI-E x1
- A300 Couger PSU (staggered spin-up ftw! I just hope it works out ;-))
- Highpoint Rocketraid 2680SGL + 2 MiniSAS to 4xSATA cables
I'll use 8 2TB HDDs for a RAID5 with Linux Ubuntu, at least that's the plan. I'll also get a UPS soon. I have enough space to upgrade to a second raid adapter (the motherboard has 2 PCI-E slots and I hope the graphics slot will be accepted) and have 15 HDDs in the 9 5.25" trays and I can cram one additional one in there for sure ;-).
Rick83 - Sunday, September 4, 2011 - link
That's a big RAID-5.Those are pretty risky: Rebuilds take a long time, and are dependant upon all other disks surviving that long.
If you really want to go the way of 'one big RAID 5', I'd propose to go with 7 disks and keep one as hot spare. That way at least the rebuild will start right as the first disk dies, minimizing somewhat that other disks deteriorate gravely until a replacement disk is there.
In general I'd stop with level 5 at 6 disks though. Consider Going with two 4-disk level 5's or a level 6 also.
Ideally of course you'd have two level 5's where one is a regular back-up of the other, but using a different file system.
qoonik - Sunday, September 4, 2011 - link
Also consider software protection like flexraid http://wiki.flexraid.com/.Death666Angel - Monday, September 5, 2011 - link
Yeah, my ideal solution was to go with RAID6 where 2 disks could fail (and of course a real RAID controller with a XOR unit and 16 sata ports) but that would have cost about 4 to 6 times the price of the 2680.I haven't heard good things about the spare disks, so I would rather go with 2 4HDD RAID5s. But I will do some testing before setting it up, I have 4 empty 2TB drives and will play around with pulling a disk out, having the array rebuilt etc. and then I'll decide which way to go.
Luckily, the only sensitive, non-easily recoverable data will be my photos and probably some system images which will be backed up regularly. The music and videos can easily be ripped from my collection again. It will be time lost, but not inrecoverable :-).
As for software RAID, I haven't heard of flexraid. I looked into FreeNAS and ZFS and that wasn't up my ally. Very powerful filesystem, but FreeNAS is too limited to just providing a NAS. I would like to have the option of going full server with this too, hosting different things. And the linux port of ZFS isn't stable as far as I heard, so that was out of the question.
With the hardware controller I know that I am not OS dependant and all that I read is that in a case of failure, modern raid controllers can also be switched easily by a model from the same maker and not lose any data (which I heard wasn't the case a few years/decades back). :-)
alpha754293 - Tuesday, September 6, 2011 - link
My current system is 10 * 3TB (30 TB raw, 27 TB RAID5). If I go with anything else, I'd have to probably pile LustreFS or some other kind of distributed FS on top of that, which adds to the complexity, and cost, and another system, and power, and complexity in the parity calculations.A second server will likely go online and it will run rsync or something akin to that for incremental backups.
Lonyo - Sunday, September 4, 2011 - link
You seem to have missed the perfect board for an Atom based file server.http://pden.zotac.com/index.php?page=shop.product_...
6 SATA ports, x16 and x1 PCIe slots (for potentially 2 RAID cards).
DTX means it should fit in an mATX case (assuming there is one with enough HDD space), or if you are making something custom the footprint shouldn't be too big.
qoonik - Sunday, September 4, 2011 - link
also suggest supermicro X7SPA-H-D525 ( Intel® Atom™ D525 (Pineview-D)Dual Core, 1.8GHz (13W) processor, 6 sata,2x RJ45 LAN port, Mini ITX)