Build a RAID 6 array for $100/TB

Build a RAID 6 array for $100/TB

Summary: Build your own cloud-style storage for $100/TB that makes Amazon's S3 look pricey. Perfect for ripping 10,000 DVDs to disk!

TOPICS: Storage, Hardware

Imagine storage that didn't cost much more than bare drives. High density storage with RAID 6 - double-fault - protection, reasonable bandwidth and web-friendly HTTPS access.

And really, really cheap.

Not your enterprise's RAID array Raw disk cost is only 5-10% of an enterprise RAID system's cost. The rest goes for corporate jets, sales commissions, tradeshows, sheetmetal, 2 Intel x86 mobos, obscene profits and some pale and blinking engineers in a windowless lab who make it work.

But what if you don't want 4-color brochures or the barely-clad booth babes. What if you just want cheap economical and reliable storage?

You aren't running the global financial system - what's left of it anyway - or a 500 person call center. But you want enough redundancy so it will stay up until morning.

Meet the Storage Pod You aren't the only one. Backblaze, a new online backup provider, designed the Storage Pod for their own use and are sharing it with everyone. They aren't selling it - that's where the build comes from - so they aren't trying to get rich off you.

Here's what it looks like:

Here's an exploded diagram with a simplified BOM:

And then there's the (free) software. 64-bit Debian Linux, IBM's open source JFS file system and HTTPS access. Simply stated each file gets a URL. Put a web server in front of it and serve the world - or just your home.

The Storage Bits take Many applications just need a big bucket that doesn't cost $5,000/TB. This is it.

You can build it yourself, but it is probably more complex than a high-end gamer system. Download the 3D SolidWorks files and have Protocase build you 1 or 500 of the boxes.

But the density is good, the performance is reasonable, the availability is decent and the price is right. This is a DC-3, not a 747. It is all you need for the right application.

And at $100/TB you can mirror all your data 2 or 3 or 4 times if you need more availability - and still be way less than half the cost of name brand arrays. Get the details from the Backblaze blog.

Comments welcome, of course. BTW, I'm trying out their free trial Mac online backup - at least one of the founders worked at Apple - and I'll let you know how it goes. I don't have a business relationship with them either, in case you're wondering.

Topics: Storage, Hardware

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • Is that per terabyte, or

    per 1 trillion bytes. Sorry, I just had to.
  • I want one!

    This would be the ideal replacement for my pair of MediaSmart servers and external 8 drive enclosure.
  • RE: Build a RAID 6 array for $100/TB

    Why not OpenSolaris and ZFS? Both for free download & use.

    OpenSolaris has kernel CIFS, too, so this would be a heckuva NAS unit.

  • Needs more cowbell ! /eom

  • At last

    Real design work ... instead of DROBO and HP Mediasmart WHS corporate monetising mediocrity.
  • What monster....

    I like it!!
  • Where can I get the case built?

    Damn I would love that just to have.
    • RE: Where can I get the case built?

      The company that makes the cases was mentioned in the article. It's in the second paragraph of the "The Storage Bits take" section, but here is the link as it wasn't given -
  • RE: Build a RAID 6 array for $100/TB

    Good analogy. This is a DC-3 that you put together yourself, having had no previous experience in building or flying aircraft, and your construction components are rubber bands, aluminum foil, and bailing wire. Once you've got the sucker in the air then you're going to figure out that you've got to change the engines in mid-flight and add the landing gear that you forgot to put on before take-off.

    For years now, Curtis Preston (author of the O'Reilly book "Using SANs and NAS") has said:

    Raw disks are cheap. Storage is expensive.

    To people who don't understand that statement, I would add the analogy:

    Raw carbon is cheap. Diamonds are expensive.

    Alternatively, let's start with rocks and sand, and try to figure out how we're going to turn that into tungsten filament and glass, so that we can try many thousands of different ways *not* to make a lightbulb.

    That's what Edison did. Do you want to do the same, or do you want to take the fruits of his labor and apply that lightbulb (or some other light-generating source) to some other activity?
    • Depends on how big the diamond is


      Google demonstrated years ago that "storage" doesn't have to be
      expensive. Numerous other companies have extended that insight to
      the rest of us. Blazeback continues that fine tradition.

      Your comments remind me of techies who derided PCs and NetWare
      because they weren't up to the standards of minicomputers and
      mainframes. How did that work out?

      Storage is, without a doubt, the hardest part of computing, because it
      is the only piece with a long-term anti-entropic mission. Storage is
      fighting the Universe, and the Universe has a lot of resources.

      Just as for the last 10 years you could buy a powerful multi-core
      server and run (some) free enterprise class apps on a free OS for very
      little money, that day is here for storage. Backblaze is using their
      storage in production and it works in their application.

      It isn't for every app, but as data cools there are more and more apps
      that need the capacity without the cost. Home and soho users are
      definitely there today.


      Robin Harris
      • I have an idea, I think it can be used for evil though.

        These things look kind've modular. What if an entity/company were to build these and have them distributed across the US/World. They could pay Billie-Sue Homemaker $100/month to plug it into their high speed connection.

        Imagine a scenario where an enterprising person could sell anonymous storage space to the mafia or malware company.

        They could guarantee that the data would be encrypted and distributed across several pods. The people housing the pods would never have to know what was being stored on them. Data would be hard to track down because it would be travelling to many locations. Drawbacks would include lack of oversight for each pod and lots of latency.

        I'm imagining a distributed array of redundant storage on a large scale. The more numerous the pods, the more redundant. Hell, why doesn't our government have one of these in every post office in the country?
        • Check out Cleversafe

          They use advanced erasure codes to create secure storage on public

          Very similar to your idea.

          Robin Harris
  • RE: Build a RAID 6 array for $100/TB

    I am not quite sure how you arrived at the 100/TB number I can attest that, the concept works. Minus the actual drive cost at 219 per 2tb drive, the cost I charged was about 225 per TB. Still better than the 5000 per TB which I have seen in the past. This rig also included a hot spare and the capability/possibility to grow.
  • I would do it with OpenSolaris or FreeBSD

    and use the ZFS filesystem. It's vastly superior to RAID-6 and, since we're already working with software RAID, doesn't add any extra overhead. And it's expandable, simply by replacing drives.
    • Maybe it would work now.

      I understand there were problems with OpenSolaris support of the Silicon
      Image chips.

      Robin Harris
  • disk scrubbing, encrypted storage

    It looks like they are using software raid, and not a hardware solution that includes regular disk scrubbing. Also, the data are stored unencrypted (decrypted after transfer).

    So it seems not very secure to me in two aspects:
    1. Your confidential data could be read by 3rd parties; I would prefer both encrypted transfer and encrypted storage.
    2. Without disk scrubbing you will get bit errors sooner or later tha can go undetected for a long time. Even raid-6 may run into lost data after a while. ZFS inlcudes mechanisms against such silent data loss, but JFS does not (afaik). Did they implement some proprietary mechanism to protect against this?