Hard Drive Reliability Update – Sep 2014

I got into building PCs on the side in about '96 or so and did break/fix support through college - it always seemed (I didn't keep records - I can't quantify) that Seagates accounted for more than their share of the "it won't start and is making a clicking noise" cases. So much so, that I have recommended against Seagates and towards WDs for as long as I can remember (15+yrs).

I've only had a small amount of direct experience with Hitachi drives, so I can't speak to them too much.. Though I do have a Hitachi in an enclosure that's been kicking around since about 2004, so I guess that says something...

cox proportional hazards models and KM survival curves are the big kahuna with a data set like this. basically, my impression is that you'd want to pretend that you're doing a cohort study and essentially analyze it like a big clinical trial.

and re: graphing, since you have sub-samples of the different drive models and relatively different but still small numbers of models for each brand, the big take-away graph comparing manufacturers leaves out a lot of useful nuances that could be pulled out of the table, e.g. that most of the hitachi drives are newer and you have fewer of them. it also doesn't portray how consistent failure rates are across models produced by the same brand might be, and it looks like there's a significant range of failure rates across different seagate models. so even just seeing IQRs of pooled failure rates for each manufacturer in box plots might be eye opening..

immortal time bias may be something to consider here as well...when you have some sample groups that mostly include newer individual drives that you have not yet had for a year on average, subtle differences in how the failure event is described can make a big difference in the conclusions you draw...especially in terms of uncertainty. if i have 10,000 hitachi drives and 5 of them fail in the first 6 months, the robustness of conclusions i can draw from those data are different in some important ways from similar insights drawn from a sample of 1,000 hitachi drives i've used for 5 years.

it's also not clear to me how you've dealt with replacement drives. based on what i gather from the post (and i could be totally wrong), if you have a bunch of one model of a drive failing and then get replacements for them... some might argue that refurbed drives should be analyzed almost like a separate drive model, since they often have physical differences compared to those bought via retail channels.

i'd be quite curious to dig into these data a bit further if you're willing to post the original data set...

thanks again for posting information that's quite useful and interesting

Except that also only works "most of the time".

Anyone who works at scale with drives and (RAID-)Controllers knows that even "enterprise" drives can and do take entire controllers down.

It's not uncommon to lose a full set of daisy chained JBODs to a single disk acting funny.

Consequently, and since storage clusters have to be redundant at the node-level anyway, it makes a lot of sense to skip the markup for enterprise firmwares and instead design for quick node failure detection and ejection (short timeouts).

I recently examined a set of Crucial m4s which were at 130% of usage. There was no lost data, but write bandwidth was hilariously bad (around 10-20MB/s).