Comments on: Looking For Network/Direct Attached Storage http://conigliaro.org/2007/11/25/looking-for-networkdirect-attached-storage/ cat /dev/brain | grep technology >> blog Wed, 01 Sep 2010 20:51:38 +0000 hourly 1 http://wordpress.org/?v=3.0.1 By: Brandon http://conigliaro.org/2007/11/25/looking-for-networkdirect-attached-storage/comment-page-1/#comment-86 Brandon Sat, 19 Jul 2008 18:22:09 +0000 http://www.gyrate.org/2007/11/25/looking-for-networkdirect-attached-storage/#comment-86 I've been thinking about this a lot lately. The problem with these bloated solutions is that they are way too expensive for what they are. A raid controller, network card, perhaps a custom board for managing a few hd state leds, an almost silly administration interface that never has all the features you want, and an enclosure. At first I didn't see why it was necessary to spend so much more for simply taking an operating system out of the picture. But on the other side, an os requires a bit more in the way of hardware and configuration, which is why most people would prefer to just throw money at the problem. So what to do? Spend an assload on someone else's plug and play solution, or build a fileserver and call it a day? I think neither. Personally, I'd rather develop my own solution. The problem is, essentially, managing a stack of raided drives(backups, notification of drive space limits and failures), connecting them to a network, having a good set of tools to set permissions and the like, and not breaking the bank. An enclosure, stripped down hardware, a lamp install, and a bit of time is all that should be necessary. And more likely than not, even with money lost on a raid card or motherboard that doesn't work out, the price will more than likely end up under 1k. We'll see. I’ve been thinking about this a lot lately. The problem with these bloated solutions is that they are way too expensive for what they are. A raid controller, network card, perhaps a custom board for managing a few hd state leds, an almost silly administration interface that never has all the features you want, and an enclosure. At first I didn’t see why it was necessary to spend so much more for simply taking an operating system out of the picture.

But on the other side, an os requires a bit more in the way of hardware and configuration, which is why most people would prefer to just throw money at the problem. So what to do? Spend an assload on someone else’s plug and play solution, or build a fileserver and call it a day?

I think neither. Personally, I’d rather develop my own solution. The problem is, essentially, managing a stack of raided drives(backups, notification of drive space limits and failures), connecting them to a network, having a good set of tools to set permissions and the like, and not breaking the bank. An enclosure, stripped down hardware, a lamp install, and a bit of time is all that should be necessary. And more likely than not, even with money lost on a raid card or motherboard that doesn’t work out, the price will more than likely end up under 1k. We’ll see.

]]>