Hacker News Reader

Comment by gruez | Hacker News Reader

Keep in mind there's no WD drives in the comparison, only HGST (reputed to be high quality), and tosiba (low sample count).

View on HN

I've been a long time IBM/Hitachi/HGST fan, and agree with the "high quality" assessment. Nearly every time I've used Seagate I've had problems. There was one batch I put in a home storage server that I had no problems with, otherwise they seem to fail a lot. Over the decades I've probably had close to a thousand Ultra/Desk Star drives (personal and business).

We recently replaced all the 15K Seagate drives in our dev/stg 6 year old server with Samsung 860 EVOs, because I was replacing 1-2 a month (out of ~20 disks).

I've also had good luck with Toshiba disks, though I have a much smaller sample size.

I've had a similar experience with Seagate - the usual failure is bad-blocks or a huge number of G-list remaps while SMART claims the drive is fine. Also had a Barracuda 7200 with the bricking firmware bug and a 7200 which suffered spindle-lock on a test bench and launched itself off the bench, smashing the mounts and cables in the process and came to rest in my lap.

I could deal with their quality issues if their RMA process was at least reasonable. With WDC it's a credit card number (if you want an advance RMA) and a web form to request an RMA. With Seagate it's a game to try and buy "official" packaging, a tooth-and-nail fight for an RMA number and a long wait.

They actually refused to RMA the spindle-locked drive because the PCB was damaged (well, no duh). Thankfully the supplier was more lenient.

I've had highly divergent experience with WD drives.

My home system using WD Reds has been great. It is larger than a normal home storage box (24 drives), and it gets substantially more use than normal (backing store for a bunch of VMs, mostly). I've replaced one drive in five and a half years.

Contrast with storage servers built for work with WD Golds. Those are 45 drives each and used for database backup workloads. We replace about a drive a quarter (of the 90 total), and this has been pretty consistent for a bit over two years.

Not sure if I got great Reds or if Golds are just terrible, but that's what I've seen.

45 is almost twice that of 24, and WD Gold are 7200 rpm vs 5400 rpm for the WD Reds. So could be the Golds generate too much vibration for their own good?

Hmm, I've had pretty good luck returning Seagates. Which I'm thankful for, because I have to do it a lot. I pay the few bucks for the advance replacement, something like $7-$15 sticks in my head, and use the packaging that they send the new drive out in to return it.

I've been following these Backblaze reports for a few years now, and Seagate always seemed to do worse than other brands. HGST has consistently been at the top, rivaled only by Toshiba lately. No wonder Backblaze has replaced most of their inventory with these two brands.

Disclaimer: I work at Backblaze.

> No wonder Backblaze has replaced most of their inventory with these two brands.

We Reed-Solomon encode each file across multiple drives in multiple machines, and we always use "enough parity" that the failure rate won't result in data loss.

So for us, we have a (pretty simple) little spreadsheet that takes into account drive failure rate as a cost, along with drive density (more dense drives means renting less data center space) and we let the spreadsheet kick out which drives to buy based on the cheapest total cost of ownership. Honestly, we're not brand loyal AT ALL, and we're not afraid of a higher failure rate (other than that raises the cost because we have to buy replacements for failed drives).

With that said, if an individual purchases ONE DRIVE they might value a lower drive failure rate differently than Backblaze does. But honestly, if there is a 1% drive failure rate per year or a 10% drive failure rate in a year, you should still backup the data so you can sleep at night.

Yev here -> Yea, the HGST drives continue to do really well for us, but at around 2% the Seagates are no slouches - especially when you consider the sample size. That's a pretty darn good ROI from drives that also tend to be a bit less expensive!

Yep, I suppose that's why you continue to use Seagate drives. They're often more than 2% cheaper than the competition!

I think I saw double-digit failure rates from certain Seagate models a few years ago, though :(

Yea they went through a bit of a rough patch, but it's honestly been years since that was the case, they're more in-line with other manufacturers now and the pricing is very competitive!

Reply on news.ycombinator.com