The best blog series going. Great technical writeups that I wish more companies would do - we've been doing it at our small business and customers really get a lot out of it. Also helps with marketing.
My simple rule still stands even after 20 years: Avoid Seagate.
Close to 10 years ago, I bought a "new" hard drive on Amazon. When I got it, I happened to notice some suspicious signs of wear, like corrosion on the PCB from someone's greasy fingerprint. I dive into the SMART stats and it's completely zeroed. Not quite what I'd expect from a new drive.
So then I start scanning the drive. The partition table was deleted, but it was full of data, most of it encrypted. It was also badly fragmented, so it was likely in some type of array. What I could recover in cleartext implied that it had been spun up for thousands of hours, in a Backblaze datacenter. It wasn't conclusive enough to go to Backblaze about it, so I just returned it to Amazon. I probably did a zero pass on it first, can't remember.
Encrypted or no, if that was a backblaze drive, they were disposing of drives with customer data still on them. I'm not surprised someone tried to pass it off as new on Amazon, that scam is old hat. I was very shocked to see the data still intact though.
Surprised by the number of drives with 0 failures, though it seems not all of the drives were run for the required time to qualify for the rating.
> In Q3, six different drive models managed to have zero drive failures during the quarter. But only the 6TB Seagate, noted above, had over 50,000 drive days, our minimum standard for ensuring we have enough data to make the AFR plausible.
I wonder if they have special hardware recycling arrangements with their vendors for decommissioned drives to reduce their footprint. How many magnets are laying around the office? :)
> The average age of the retired drives was just over eight years
Didn't realize they can last so long.
As much as I appreciate these articles... Guys... Please don't save up on ink! Eyes are more expensive than black pixel paint these days.
Cool article! Looks like HGST (formally Hitachi) perform very well overall. Toshiba 4GB ran 101 months with no failures, holy moly!
I used to like Samsung hard drives personally, Lacie used them in their rugged series and I found them to have pretty low failure rates. Seagate bought out Samsung's disk drive business in 2011 apparently. I guess Samsung saw the future of SSD's?
Is there any backup software which continuously uploads your files to an AWS/GCP/Azure long term storage account that you control and pay for? Something like CrashPlan, which from time to time performs automatic maintenance and deletes old versions of files.
> this chart is the confidence interval, which is the difference between the low and high AFR confidence levels calculated at 95%.
How to calculate low and high AFR ?
Backblaze is interesting but it's not very easy to use. Its interface is rather basic and it was difficult to select which drives to back up and which not to. It kept trying to back up directories on my computer that I specifically told it not to, and there was no way to efficiently update program's behaviouir from the the "what to backup list". It might be nice if you just want to backup your computer and all your drives but the moment you want only parts of your computer backed up, it's frustrating.
I run a bunch of WUH721414ALE6L4 and WUH721414ALE604 in numerous RAID10 volumes. Haven't had a failure yet.
L = without power disable, 0 = with
Kinda low sample sizes for 0 and highest AFRs
> ambient temperature within a data center often increases during the summer months
What? Isn't your Data Center supposed to be temperature controlled, where the A/C has a setpoint which it's keeping the entire environment to within a degree?
Being able to tell what season it is based on your HDD SMART temperature (armchair expert here) sounds bad.
The article discusses the impact of high absolute temperature on the longevity of drives, however from my amateur knowledge the range of temperature during a day is also an important factor.
I always assumed that having a stable 40°C is better than a drive constantly swinging between 20°C and 40°C, so I am surprised that the article only mentions alerts on reaching a high threshold.