On 08/29/2014 04:04 PM, Evan Root wrote: > It seems that after reading the backblaze and google papers about drive > reliability that there is no statistically obvious difference. It's too > close to call. Both papers end with hedges and further questions. Even if > enterprise drives are more reliable it isn't by more than a single percent > and it isn't consistent enough to matter either. > > Evan Root, CCNA > > > > On Thu, Aug 28, 2014 at 1:40 PM, Chris Cappuccio <ch...@nmedia.net> wrote: > >> Christian Weisgerber [na...@mips.inka.de] wrote: >>> Now, the real question is whether enterprise drives actually *are* >>> more reliable than consumer drives. >>> >> For regular hard disks, the answer is definitely, no. What I extract from the various verbiage is:
Under benign conditions, i.e. no significant vibration and good cooling, the two are virtually equally reliable. Under harsh conditions, specifically high vibration and high heat due to constant activity, the desktop drives are likely to fail quickly and the enterprise drives will continue to operate. The various grades of WD disks are labeled as to how many disks will be in the chassis. IIRC something like 1 or 2, up to 5, and lots for the 3 grades. This corresponds to the amount of vibration hardening. Now, with good cooling and some shock mounting which does vibration isolation but holds the drives so that they don't move during seeks (conflicting requirements, ingenuity required) desktop drives might be usably reliable. So comparing reliability statistics is only meaningful if the case, cabinet, and cooling are equivalent. My guess is that the enterprise drives are aimed at data centers which pack drives into the smallest possible space and can't afford to pull them to replace them. They would pay the 3X premium.
