[Boards: 3 / a / aco / adv / an / asp / b / bant / biz / c / can / cgl / ck / cm / co / cock / d / diy / e / fa / fap / fit / fitlit / g / gd / gif / h / hc / his / hm / hr / i / ic / int / jp / k / lgbt / lit / m / mlp / mlpol / mo / mtv / mu / n / news / o / out / outsoc / p / po / pol / qa / qst / r / r9k / s / s4s / sci / soc / sp / spa / t / tg / toy / trash / trv / tv / u / v / vg / vint / vip / vp / vr / w / wg / wsg / wsr / x / y ] [Search | Extra juicy! | Home]

Backblaze HDD Failure Rate: UPDATE

This is a blue board which means that it's for everybody (Safe For Work content only). If you see any adult content, please report it.

Thread replies: 46
Thread images: 3

https://www.backblaze.com/blog/hard-drive-reliability-q3-2015/

>le WD meme
>hurr durr don't buy seagate
>>
>>51594144
>inb4 "never failed on me"
>>
>>51594144
>>51594159
Someone doesn't understand the bathtub curve.
Some of these WD drives are over 5 years old now.
>>
>>51594144
is this not the people who bought cheap external harddrives from costo and mounted them on a piece of plywood to run as servers or storage ?
>>
I find it funny that people spend ass loads on various WD colors. I've always just bought the cheapest HGST shit and its always worked great running 24/7 in raid for years, no issues, at 2/3 the price of any WD drive.
>>
>>51594378
I've heard Toshibas are the same, but I've never had good luck with those.
>>
>>51594378
Are you buying old or used? HGST isn't cheap.
>>
File: blog-q3-stats-table.png (337KB, 681x1270px) Image search: [iqdb] [SauceNao] [Google]
blog-q3-stats-table.png
337KB, 681x1270px
>>
in before someone claims these tests are not valid because the hard drives are stressed too much or whatever gay faggot excuse they have
>>
>>51594500
Oh they're valid. The only used WD Reds and Greens though. I'm not entirely sure how Seagate grabbed a 222.77% failure rate for their ST1500DL003. It was so bad 100% of them failed so they replaced all of them and 100% of those failed and 22.77% of those failed?
>>
File: 1426182004103.png (335KB, 533x354px) Image search: [iqdb] [SauceNao] [Google]
1426182004103.png
335KB, 533x354px
>>51594485
>Seagate Barracuda Green 1.5TB 2015 Failure Rate: 222.77%
>Whenever a drive fails, it fails twice
How?
>>
>>51594411
>HGST isn't cheap
What planet have you been living on?
>>
>>51594607
Drive fails within the year, is replaced, and the new drive ALSO fails within the year.
>>
>>51594665
but the average time to failure for that drive is listed as over two years. It would be much lower if every drive-slot failed twice or more in a year
>>
>>51594712
They got duds. Happens sometimes.
>>
>>51594742
It's a calculated time to failure though. If they just happened to get 80% duds somehow, it would have driven down the failure time average
>>
>>51594628
I'm looking at the Deskstars and Ultrastars on Newegg and they are all more expensive than the cheapest WD and Seagate drives. They look cheaper than WD enterprise, but I think Seagate are similarly priced.
>>
>>51594799
>He believes the average times to fail aren't bullshit
Just how Jewed are you?

If they published the factual results from a group of defective drives NOBODY would buy them.

They more than likely take the best group of drives and derive a mean off of those.
>>
>>51594144
wasnt this backblaze shit rigged and debunked?
>>
>>51594830
Can you fucking read? These are the *calculated* times to failure. As in "calculated by backblaze using this test"
They aren't the fucking advertised failure times
>>
>>51594831
It was never a thorough test of every model. They just buy the cheapest drives on the market at the time and post failure rates. There's no reason not to believe those numbers though.
>>
>>51594831
it was rigged in the sense that the way they use hard drives is not representative of most users, and I think I remember something about them sourcing hard drives in a way that made it not a good sample (i.e. it wasn't random).

The backblaze stuff is still useful if you use caveats, the way you should with any statistical, sampled, quantitative data.

I imagine that the people who completely wrote off backblaze's data are the same sorts of people who simply don't understand how easily you can measure/quantify shit in a biased or otherwise broken way and utterly obscure that bias because we tend to trust numbers as honest.
>>
>>51594831
The three criticisms I've seen leveled are that the Seagate 3TBs were sometimes shucked drives, the older models were put in badly designed enclosures that might've accelerated failure, and that the way they report results is poor.

If all of that is taken into consideration, these reports can still be useful, especially on newer models. They're just not the law.
>>
>greens
kek'd slightly
>>
I have a bunch of Toshiba drives in my NAS that are apparently rebranded Hitachi drives or some shit? Thank god they aren't Seagate or WD at least.
>>
What is it about WD Reds that make them almost as bad as WD Greens? I don't think they have the aggressive head parking issues.
>>
>>51594144
So I assume the Hitachi Deathstar meme is not particularly accurate then.
>>
>>51597009
Isn't that just the 75GXP?
>>
>>51594144

Oh, its another test from the company that uses consumer drives in NAS enviorments and constant load? That must be 100% accurate.
>>
>>51597181
It's almost as if the results are completely random
>>
>>51594267
>Some of these WD drives are over 5 years old now.
as are some of the seagates. Whats your point?
>>
>>51594913
>There's no reason not to believe those numbers though.
They run it 24/7 in a raid environment where you might have it on 4 hours a day sometimes.The drives they are using are not meant for this kind of usage and reliability will be much different from home use.
>>
>>51597477
>Whats your point?
Perhaps you should try reading over the first part of his post instead of skipping right over it.
>>
>>51597181
Reminder that load has no effect on drive lifespan according to Google's study on it.
>>
>>51597610
I skipped over it because its is moot. Seagate has more drives in production and are just as old.
>>
>>51597477
>as are some of the seagates.

Yeah, the few survivors.

Most of the Seagates are newer than most of the WD's because Seagate's failure right was so appallingly high.

Anyway

>implying I'm not running HGST everywhere

I have some WD USB drives and they're great. It's all HGST + Sandisk SSD otherwise.
>>
>>51594144
AFAIU, in consumer electronics, the failure rates depend A LOT on the factors like:
- quality of the components
- quality of the materials
- quality of the production process
the very same thing happens with RAM modules.
buying something without researching about it is practically wishful thinking: you hope and wish that the (whole series of) product you bought is a lucky, good one.
you don't, ever, trust brands in technology, they all play with the laws of physics, and they sometimes lose to the competition.
>>
>>51594485
>222%
>>
>>51597181
Reminder that enterprise drives are no different.[1]

[1]: https://www.backblaze.com/blog/enterprise-drive-reliability/
>>
Takeaway: There are a few known-bad models of drive that you have to avoid. Other than that, you'll probably be fine whatever you get.
>>
>>51594144
Yeah, still avoid Seagate. Why would you purposely buy from the seemingly second worst brand?

Just stop being a faggot and go HGST.
>>
>>51594817
http://www.newegg.com/Product/Product.aspx?Item=N82E16822146129&cm_re=2tb_ultrastar-_-22-146-129-_-Product
It's not the newest, but still.
>>
>>51600218
>wants reliability
>buys multi-platter

Doing it wrong.
>>
>>51602603
I'm merely saying if he believes that chart so much why settle for the second worst.
>>
>>51594159
Nigga i have 6 external WD drives of a total of 5TB. Last one was bought 2 years ago, first one i dont even remember. Never failed
Oh i also have 1 Toshiba 500GB. Guess what. Failed

I am not a first-time buyer so i dont have to resort to trusting a random graph. Personal positive feedback > dubious research
>>
>>51600218
Why would you avoid the best 1TB that also happens to be the cheapest simply because their previous gen 1.5TBs and latest 3TBs were utter crap?

http://hdd.userbenchmark.com/Seagate-Barracuda-720014-1TB/Rating/1849
Thread posts: 46
Thread images: 3


[Boards: 3 / a / aco / adv / an / asp / b / bant / biz / c / can / cgl / ck / cm / co / cock / d / diy / e / fa / fap / fit / fitlit / g / gd / gif / h / hc / his / hm / hr / i / ic / int / jp / k / lgbt / lit / m / mlp / mlpol / mo / mtv / mu / n / news / o / out / outsoc / p / po / pol / qa / qst / r / r9k / s / s4s / sci / soc / sp / spa / t / tg / toy / trash / trv / tv / u / v / vg / vint / vip / vp / vr / w / wg / wsg / wsr / x / y] [Search | Top | Home]
Please support this website by donating Bitcoins to 16mKtbZiwW52BLkibtCr8jUg2KVUMTxVQ5
If a post contains copyrighted or illegal content, please click on that post's [Report] button and fill out a post removal request
All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.
This is a 4chan archive - all of the content originated from that site. This means that 4Archive shows an archive of their content. If you need information for a Poster - contact them.