Home
| Calendar
| Mail Lists
| List Archives
| Desktop SIG
| Hardware Hacking SIG
Wiki | Flickr | PicasaWeb | Video | Maps & Directions | Installfests | Keysignings Linux Cafe | Meeting Notes | Linux Links | Bling | About BLU |
> From: discuss-bounces+blu=nedharvey.com at blu.org [mailto:discuss- > bounces+blu=nedharvey.com at blu.org] On Behalf Of Tom Metro > > It would be interesting to get an expert's opinion on this, backed up > with some long term statistics, to see if drive reliability, at the > higher capacities, is in fact trailing off. def longwindedreply: print """ I can say that the BER of the actual media is definitely increasing with increased density and decreased cost (especially in flash) - But you compensate for this by implementing a more robust FEC. One of the companies I work for has been advancing the FEC for the last several years. You can compensate as much or as little as you want (but of course, it costs more development dollars and time to market, in order to compensate more for high BER). So you choose a design requirement relative to some currently accepted standard, and you stop developing when you reach that target. So, based on what I just said, the reliability of the newer higher density drives is the same as the reliability of the more traditional drives (by design.) But our chips don't go into every single drive in the world... And I can't tell you what other manufacturers are doing. I can't even tell you who our customers are, thanks to NDA. ;-) One would hope, that when you buy a more expensive drive labeled "enterprise," that you're actually paying for something - vis-a-vis good engineering & manufacturing. But there is really no way to be certain. As an IT guy and Computer Engineer, and EE, working for CS/EE / chip companies, I encourage companies to stick with big name brands where they've sold millions of the same drive under warranty (such as ... buy your Dell and Apple and Oracle branded drives rather than OTS seagate/westerndigital/etc drives.) But it's all just a superstition and a balance of probabilities, and in fact, buying the supported name branded drives does indeed cost more per drive. Here's how I make my own choices: In really big important servers, I use the name branded officially supported drives. Otherwise, I use the cheaper OTS drives. I have a bias in favor of Seagate, just because I've experienced fewer failures and greater ease of warranty exchange on these drives relative to their competitors. And as mentioned before in this thread, I have a bias away from Green or Environmentally friendly or cheap drives. I always look at the length of the Mfgr warranty, and nothing less than 3-5 yrs is satisfactory. If I buy the Dell/Oracle drives, I know I will have the opportunity to extend the warranty after 3 yrs if I care to. With apple, it's not an option. With commodity OTS drives, it's also not an option. So in the former case, I buy just how many I need, and with the latter case, I pre-stock twice as many drives because I know 3 yrs from now, these things won't be for sale anymore. So ... If the commodity OTS drives cost half as much as the big-name-branded drives and I buy twice as many of them ... It's basically a wash. :-) """
BLU is a member of BostonUserGroups | |
We also thank MIT for the use of their facilities. |