Personal computing discussed

Moderators: morphine, Steel

 
CScottG
Gerbil Elite
Posts: 855
Joined: Fri Dec 01, 2006 9:53 pm

Re: I need some help, gerbils! Which are the quiet consumer hard drives?

Tue Aug 08, 2017 9:06 pm

Yup, that Seagate is pretty good capacity for the dollar (for 2.5"), including the smaller 4tb version. Definitely a better purchase in the US than the UK.. :oops:
 
Chrispy_
Gold subscriber
Grand Gerbil Poohbah
Topic Author
Posts: 3958
Joined: Fri Apr 09, 2004 3:49 pm
Location: Europe, most frequently London.

Re: I need some help, gerbils! Which are the quiet consumer hard drives?

Wed Aug 09, 2017 6:26 am

Thanks for all the feedback guys.

I've ordered a trio of WD Red 8TB drives to give me 16TB usable in an array. On reflection it's pretty expensive at £750, given that I could shuck a couple of external 8TB drives for under £350 but this way I get the peace and quiet I want (hopefully), and the protection of RAID5 for at least one drive failure between my fortnightly backups.

I've high hopes for the noise levels, given that these are based on the Ultrastar He8 drives with a reduced spindle speed and the enterprise drives at 7200rpm are pretty quiet already.

I just feel a bit dirty giving WD my money (again), but I don't owe Seagate or Toshiba anything either; It's their own fault for not having a 5400/5900rpm product at the capacity I'm after.
¯\_(ツ)_/¯
Congratulations, you've noticed that this year's signature is based on outdated internet memes; CLICK HERE NOW to experience this unforgettable phenomenon. This sentence is just filler and as irrelevant as my signature.
 
just brew it!
Gold subscriber
Administrator
Posts: 48532
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: I need some help, gerbils! Which are the quiet consumer hard drives?

Wed Aug 09, 2017 7:41 am

RAID 5 is deprecated for newer high capacity drives, due to the non-trivial risk of a second drive failure during an array rebuild (since the rebuild takes so long and involves reading so much data).
If the world isn't making sense to you, you're either drinking too much or not drinking enough.
 
Topinio
Graphmaster Gerbil
Posts: 1213
Joined: Mon Jan 12, 2015 9:28 am
Location: London

Re: I need some help, gerbils! Which are the quiet consumer hard drives?

Wed Aug 09, 2017 8:47 am

just brew it! wrote:
RAID 5 is deprecated for newer high capacity drives, due to the non-trivial risk of a second drive failure during an array rebuild (since the rebuild takes so long and involves reading so much data).
That's been true for a decade. By my maths and the specs, rebuilding this array has a theoretical 72.28% chance of failure. Assuming there's a hot spare in the chassis.

( 16 TB needs reading, 128 Tb, 1.28E14 b. Probability of not having a bit be unreadable is (1-(1/UBER)) per bit, and as each bit read is an independant event, it's exponential to number of bits. With a consumer UBER of 1E14 and needing to read 1.28E14 bits, (1-(1/1E14))**1.28E14 = 72.17% chance of an uncorrectable read error on that degraded array. With RAID-5, a URE during the rebuild process is catastrophic.

Chance of one of the other drives failing during the rebuild is much smaller but non-negligible. At an assumed 130 MB/s, rebuild time is 34h. A MTBF of 1E6 h is an AFR of 8.7E-1 which translates into an hourly failure risk of a bit over 1 in 1E6, a daily failure risk of 0.0024%. 34h has a failure risk of 0.0034% per drive, which is a failure risk of 0.12% for either of the pair.

72.17+0.12 = 72.28% ; if it takes 5 d to get a spare that's up to 74.50% &c.)

I don't think this is wrong, sure someone will disagree though.
Desktop: E3-1270 v5, X11SAT-F, 32GB, R9 290X, 250GB BX100, 2TB Ultrastar, Xonar DGX, XL2730Z
HTPC: i5-2500K, DH67GD, 6GB, GT 1030 SC, 250GB BX100, 1.5TB Barracuda, Xonar DX
 
morphine
Gold subscriber
Gerbilus Supremus
Posts: 11341
Joined: Fri Dec 27, 2002 8:51 pm
Location: Portugal (that's next to Spain)

Re: I need some help, gerbils! Which are the quiet consumer hard drives?

Wed Aug 09, 2017 9:05 am

Chrispy, I would have advised you against drives so big on the basis that if one of them fails out of warranty and you have to buy a new one, you're stuck with a rather large bill. Then again, if you're that gung-ho about silence, you want as fewer drives as possible, so hey, whatever works for you.

I went through that just two weeks ago, an old 4TB Green I had in my mostly-5TB array gave up the ghost, and I had to buy a 6TB drive to replace it since I couldn't find 5TB drives anymore. I suddenly realized that if I were to do this all over again, I'd have used 4TB spinners at most to ease the pain of buying a replacement, you can get those for dirt cheap now.
There is a fixed amount of intelligence on the planet, and the population keeps growing :(
 
just brew it!
Gold subscriber
Administrator
Posts: 48532
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: I need some help, gerbils! Which are the quiet consumer hard drives?

Wed Aug 09, 2017 9:29 am

Topinio wrote:
just brew it! wrote:
RAID 5 is deprecated for newer high capacity drives, due to the non-trivial risk of a second drive failure during an array rebuild (since the rebuild takes so long and involves reading so much data).
That's been true for a decade. By my maths and the specs, rebuilding this array has a theoretical 72.28% chance of failure. Assuming there's a hot spare in the chassis.

( 16 TB needs reading, 128 Tb, 1.28E14 b. Probability of not having a bit be unreadable is (1-(1/UBER)) per bit, and as each bit read is an independant event, it's exponential to number of bits. With a consumer UBER of 1E14 and needing to read 1.28E14 bits, (1-(1/1E14))**1.28E14 = 72.17% chance of an uncorrectable read error on that degraded array. With RAID-5, a URE during the rebuild process is catastrophic.

Chance of one of the other drives failing during the rebuild is much smaller but non-negligible. At an assumed 130 MB/s, rebuild time is 34h. A MTBF of 1E6 h is an AFR of 8.7E-1 which translates into an hourly failure risk of a bit over 1 in 1E6, a daily failure risk of 0.0024%. 34h has a failure risk of 0.0034% per drive, which is a failure risk of 0.12% for either of the pair.

72.17+0.12 = 72.28% ; if it takes 5 d to get a spare that's up to 74.50% &c.)

I don't think this is wrong, sure someone will disagree though.

I believe your analysis is more or less correct from a theoretical standpoint.

Where it diverges from reality is that the actual uncorrected error rates we see from real drives tends to be substantially lower than the UBER would imply. So the probability of a rebuild failure is lower in practice, but still non-trivial.

My recommendation in this case would've been to go with a RAID-6 array of smaller devices.
If the world isn't making sense to you, you're either drinking too much or not drinking enough.
 
Topinio
Graphmaster Gerbil
Posts: 1213
Joined: Mon Jan 12, 2015 9:28 am
Location: London

Re: I need some help, gerbils! Which are the quiet consumer hard drives?

Wed Aug 09, 2017 9:53 am

just brew it! wrote:
I believe your analysis is more or less correct from a theoretical standpoint.

Where it diverges from reality is that the actual uncorrected error rates we see from real drives tends to be substantially lower than the UBER would imply. So the probability of a rebuild failure is lower in practice, but still non-trivial.

My recommendation in this case would've been to go with a RAID-6 array of smaller devices.

Yes on the IRL BER, but if it were 1E-15 the vendors would put that because they do on those drives which can do it.

1E-15 drives in this set-up implies 12% not 72% chance of URE on rebuild, so that's a window of 12-72% ...

On RAID-6, it has the same UBER so it's the same chance of a read-failure. If a drive goes in a 4-drive RAID-6 of these, you still need to read 1.28E14 bits to write the new disk's 6.4E13 bits. Still a 12% to 72% chance you can't do so -- you've just got another disk to try from when the URE happens.

Edit: If we go for a first approximation that the parity means you have double your chances at reading a bit (independance of the parity and data) and work on a basis that there is a RAID controller which will mark the unreadable bit bad and remap unused space ( :-? ) then with any 16 TB usable RAID-6 with disks at 1E-14 I make it only a 53% chance of being able to read the array to rebuild. It's a 6% chance if the UBER is 1E-15.
Desktop: E3-1270 v5, X11SAT-F, 32GB, R9 290X, 250GB BX100, 2TB Ultrastar, Xonar DGX, XL2730Z
HTPC: i5-2500K, DH67GD, 6GB, GT 1030 SC, 250GB BX100, 1.5TB Barracuda, Xonar DX
 
Waco
Gold subscriber
Gerbil Jedi
Posts: 1918
Joined: Tue Jan 20, 2009 4:14 pm
Location: Los Alamos, NM

Re: I need some help, gerbils! Which are the quiet consumer hard drives?

Wed Aug 09, 2017 9:58 pm

Topinio wrote:
just brew it! wrote:
I believe your analysis is more or less correct from a theoretical standpoint.

Where it diverges from reality is that the actual uncorrected error rates we see from real drives tends to be substantially lower than the UBER would imply. So the probability of a rebuild failure is lower in practice, but still non-trivial.

My recommendation in this case would've been to go with a RAID-6 array of smaller devices.

Yes on the IRL BER, but if it were 1E-15 the vendors would put that because they do on those drives which can do it.

1E-15 drives in this set-up implies 12% not 72% chance of URE on rebuild, so that's a window of 12-72% ...

On RAID-6, it has the same UBER so it's the same chance of a read-failure. If a drive goes in a 4-drive RAID-6 of these, you still need to read 1.28E14 bits to write the new disk's 6.4E13 bits. Still a 12% to 72% chance you can't do so -- you've just got another disk to try from when the URE happens.

Edit: If we go for a first approximation that the parity means you have double your chances at reading a bit (independance of the parity and data) and work on a basis that there is a RAID controller which will mark the unreadable bit bad and remap unused space ( :-? ) then with any 16 TB usable RAID-6 with disks at 1E-14 I make it only a 53% chance of being able to read the array to rebuild. It's a 6% chance if the UBER is 1E-15.


That's the theory, but like I said, in practice the numbers for UBER are *far* lower incidence than rated (I wish I could quantify this further, but much of that data is NDA). Further, if one hits, it's not like you're screwed...you just try the rebuild again.

Anecdote: I have a set of old 18+2 RAID 6 arrays of 2 TB drives. They've been around forever, and they're only around now out of convenience (losing them wouldn't hurt anything at all). Out of the 8 arrays, they've rebuilt many tens of times in the past 4 years (at least 50 times, probably close to 100). Not a single one ever failed during rebuild, and these are populated with crappy consumer-type SATA drives.
Z170A Gaming Pro Carbon | 6700K @ 4.5 | 16 GB | GTX Titan X | Seasonix Gold 850 | XSPC RX360 | Heatkiller R3 | D5 + RP-452X2 | Cosmos II | Samsung 4K 40" | 480 + 240 + LSI 9207-8i (128x8) SSDs
 
Topinio
Graphmaster Gerbil
Posts: 1213
Joined: Mon Jan 12, 2015 9:28 am
Location: London

Re: I need some help, gerbils! Which are the quiet consumer hard drives?

Thu Aug 10, 2017 4:24 am

Waco wrote:
That's the theory, but like I said, in practice the numbers for UBER are *far* lower incidence than rated (I wish I could quantify this further, but much of that data is NDA).
Data (or information based on internal vendor datasets) should show better than spec performance when a spec is a limit, but surely if it were massively better across the entire ensemble they'd change the published spec to not be making themselves look worse than they are. Like they do with e.g. some of the drives having a MTBF of 1.2E6, while others have 1E6.

If you can't share the data, fair enough, but do you mean "far lower" as in one (or more) orders of magnitude lower (base 10) ? Or "far lower" as in the rating is "<1 in 1014" and the actual performance is 2 in 1015 which is close to the next band but not quite there ?

If the former, i.e. if the 1E-14 ones are actually 1E-15 or 1E-16 or whatever, then it seems nonsensical to me: why on earth would every drive manufacturer sell some drives spec'd as 1E-14 and others as 1E-15 ??
Desktop: E3-1270 v5, X11SAT-F, 32GB, R9 290X, 250GB BX100, 2TB Ultrastar, Xonar DGX, XL2730Z
HTPC: i5-2500K, DH67GD, 6GB, GT 1030 SC, 250GB BX100, 1.5TB Barracuda, Xonar DX
 
just brew it!
Gold subscriber
Administrator
Posts: 48532
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: I need some help, gerbils! Which are the quiet consumer hard drives?

Thu Aug 10, 2017 7:07 am

The negative effects of UBER should be mitigated by RAID systems that do a periodic scrub. Some percentage of sectors which had undetected write failures or which have gone bad after being written get detected and rewritten with good data before a drive failure forces a replacement and rebuild. I'm sure the raw UBER specs don't take this into account.
If the world isn't making sense to you, you're either drinking too much or not drinking enough.
 
Duct Tape Dude
Gold subscriber
Gerbil Elite
Posts: 553
Joined: Thu May 02, 2013 12:37 pm

Re: I need some help, gerbils! Which are the quiet consumer hard drives?

Thu Aug 10, 2017 7:13 am

Topinio wrote:
If the former, i.e. if the 1E-14 ones are actually 1E-15 or 1E-16 or whatever, then it seems nonsensical to me: why on earth would every drive manufacturer sell some drives spec'd as 1E-14 and others as 1E-15 ??
Market segmentation! Configure the firmware to have a drive retry failed reads twice as much and increase reliability by "10x" and BAM! 30% premiums for the same drives with a different color label. Pulling an Intel is free money.
 
just brew it!
Gold subscriber
Administrator
Posts: 48532
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: I need some help, gerbils! Which are the quiet consumer hard drives?

Thu Aug 10, 2017 7:35 am

Duct Tape Dude wrote:
Topinio wrote:
If the former, i.e. if the 1E-14 ones are actually 1E-15 or 1E-16 or whatever, then it seems nonsensical to me: why on earth would every drive manufacturer sell some drives spec'd as 1E-14 and others as 1E-15 ??

Market segmentation! Configure the firmware to have a drive retry failed reads twice as much and increase reliability by "10x" and BAM! 30% premiums for the same drives with a different color label. Pulling an Intel is free money.

UBER isn't a "hard" number like sustained transfer rate or seek time (which are inherent to the design of the drive). There's inevitably going to be variation in the manufacturing process of the various components which will affect overall error rate. If they're being at all honest, the manufacturing and QA process will be "tuned" to ensure that the vast majority of the drives shipped meet the published UBER number. Assuming some sort of statistical distribution (bell curve?), most of the drives will be significantly better.

While I'm sure segmentation plays a role, it's not the end of the story. Enterprise drives with better UBER numbers also tend to have shorter retry timeouts, on the assumption that they are more likely to be used in a RAID array where you want to report failed blocks quickly, and let the RAID controller deal with it. They also tend to have features like tighter QA on spindle balance (to reduce wear and prevent vibration from affecting adjacent drives in dense servers), more robust mechanical design (e.g. some high-end drives support the spindle shaft at both ends instead just of one), and/or exotic design features like a sealed helium-filled chamber (to facilitate increased density and reduce friction).
If the world isn't making sense to you, you're either drinking too much or not drinking enough.
 
Waco
Gold subscriber
Gerbil Jedi
Posts: 1918
Joined: Tue Jan 20, 2009 4:14 pm
Location: Los Alamos, NM

Re: I need some help, gerbils! Which are the quiet consumer hard drives?

Thu Aug 10, 2017 8:00 am

just brew it! wrote:
The negative effects of UBER should be mitigated by RAID systems that do a periodic scrub. Some percentage of sectors which had undetected write failures or which have gone bad after being written get detected and rewritten with good data before a drive failure forces a replacement and rebuild. I'm sure the raw UBER specs don't take this into account.


This also helps, but the spec is still wildly conservative IMO.

Especially now with ZFS making detecting and reporting these errors trivial...I have many tens of petabytes of storage that has been read through many times without encountering a single UBE. If the spec was even within an order or magnitude or two, surely we'd have seen something...and these aren't enterprise dives I'm talking about.

Someone should contact Backblaze and see if they monitor this stuff. :)
Z170A Gaming Pro Carbon | 6700K @ 4.5 | 16 GB | GTX Titan X | Seasonix Gold 850 | XSPC RX360 | Heatkiller R3 | D5 + RP-452X2 | Cosmos II | Samsung 4K 40" | 480 + 240 + LSI 9207-8i (128x8) SSDs
 
Vhalidictes
Graphmaster Gerbil
Posts: 1386
Joined: Fri Jan 07, 2005 2:32 pm
Location: Paragon City, RI

Re: I need some help, gerbils! Which are the quiet consumer hard drives?

Thu Aug 10, 2017 11:31 am

Krogoth wrote:
Modern HDDs are pretty quiet for the most part. They have been that way ever since they moved to fluid bearings. You get better luck with using noise damping kits/mounts if you feel that modern HDDs are too loud for your tastes.


This. Avoid Toshiba/ "true 7200 RPM" drives and you'll be fine for noise. Those are the only really loud ones I've encountered.

As far as reliability goes... nothing is reliable. There, I've said it. Anything over 1TB and you're going to have a horrible failure rate, it's all RNG.
 
just brew it!
Gold subscriber
Administrator
Posts: 48532
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: I need some help, gerbils! Which are the quiet consumer hard drives?

Thu Aug 10, 2017 11:51 am

Vhalidictes wrote:
As far as reliability goes... nothing is reliable. There, I've said it. Anything over 1TB and you're going to have a horrible failure rate, it's all RNG.

I'm going to disagree slightly; there are more reliable models, and less reliable ones, so you can theoretically bias the RNG in your favor. But in the absence of additional data (e.g., you buy new old stock of a model that Backblaze beat on for months and proclaimed more reliable than average), I'd agree that modern drives are a bit of a crapshoot.

RAID plus a good backup strategy is the only reliable way to protect yourself from downtime and data loss.
If the world isn't making sense to you, you're either drinking too much or not drinking enough.
 
HERETIC
Gerbil XP
Posts: 352
Joined: Sun Aug 24, 2014 4:10 am

Re: I need some help, gerbils! Which are the quiet consumer hard drives?

Sun Aug 13, 2017 4:00 am

Vhalidictes wrote:

As far as reliability goes... nothing is reliable. There, I've said it. Anything over 1TB and you're going to have a horrible failure rate, it's all RNG.


Agree that spinning rust can be a cr@pshoot/luck of the draw....................
Don't agree with your 1TB limit.
My most reliable/longest serving are some 2TB-HD204-Samsung drives.
 
just brew it!
Gold subscriber
Administrator
Posts: 48532
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: I need some help, gerbils! Which are the quiet consumer hard drives?

Sun Aug 13, 2017 11:18 am

HERETIC wrote:
Vhalidictes wrote:

As far as reliability goes... nothing is reliable. There, I've said it. Anything over 1TB and you're going to have a horrible failure rate, it's all RNG.

Agree that spinning rust can be a cr@pshoot/luck of the draw....................
Don't agree with your 1TB limit.
My most reliable/longest serving are some 2TB-HD204-Samsung drives.

And I had some sub-1TB Seagates that did not work out well.
If the world isn't making sense to you, you're either drinking too much or not drinking enough.
 
Bauxite
Gerbil Elite
Posts: 709
Joined: Sat Jan 28, 2006 12:10 pm
Location: electrolytic redox smelting plant

buy once cry once

Sun Aug 13, 2017 12:58 pm

Shucked 8x8TB reds last year for $225/drive (duo deal), shucked another 8x8TB last month for $160/drive @ BB, rolled all thailands, good stuff.

Friends don't let friends RAID 5. 6/Z2+ or mirrors only.

ZFS + fast LAN + SSDs in everything else = I hear nothing.

I prefer 10 for speed (you notice when source path is SSD) and least pain to deal with. I like arrays of 6~8 to keep them on one SAS card for easy system migration, and you can split mirrors across each port for any cable problems. I also think doing weird crap like 24-bay RAID 6/Z2 that performs like a super bad latency single drive is a clear sign of a penny wise and pound foolish person.
E5 1680v2 8c16t 4.5Ghz - "Yes Virginia, there were unlocked xeons"
E5 2696v4 22c44t 2.2~3.7Ghz Because Reasons™
TR 1950x 16c32t OTW
 
Waco
Gold subscriber
Gerbil Jedi
Posts: 1918
Joined: Tue Jan 20, 2009 4:14 pm
Location: Los Alamos, NM

Re: buy once cry once

Sun Aug 13, 2017 3:48 pm

Bauxite wrote:
I also think doing weird crap like 24-bay RAID 6/Z2 that performs like a super bad latency single drive is a clear sign of a penny wise and pound foolish person.

It totally depends on your workload. For NAS duties (not random read-heavy with large datasets) a wider setup is pretty decent. For writing it almost doesn't matter at all.

24 drives in a Z2, though, is a recipe for data loss in the long run. :) Even our "wide" arrays of 20 drive Z3 arrays are protected via additional layers at work.
Z170A Gaming Pro Carbon | 6700K @ 4.5 | 16 GB | GTX Titan X | Seasonix Gold 850 | XSPC RX360 | Heatkiller R3 | D5 + RP-452X2 | Cosmos II | Samsung 4K 40" | 480 + 240 + LSI 9207-8i (128x8) SSDs
 
frumper15
Gerbil XP
Posts: 346
Joined: Mon Jan 18, 2010 3:25 pm

Re: I need some help, gerbils! Which are the quiet consumer hard drives?

Wed Aug 16, 2017 9:25 pm

So I thought of this thread today when I was installing a new 8TB Seagate drive I shucked from one of these: https://www.amazon.com/Seagate-Expansio ... B01HAPGEIE and inside it had a ST8000DM004-2CX188. The ST8000DM004 seems to be a bit of a weirdo in Seagate's product lineup - as best I can tell, it is different than the ST8000DM0004 (with 3 zeros at the end) which is a $300+ BarraCuda Pro drive running at 7200 RPM.

This document describes the 004 drive, but no mention anywhere of RPM that I found (or recording technology that I was really interested in). http://www.seagate.com/www-content/prod ... 05918d.pdf

The best I can tell is from another guy that shucked the drive and did a lot more digging than I have had time for.
https://forums.lime-technology.com/topi ... ive-drive/

As best I can tell, I've got a 5400 RPM drive that uses standard perpendicular recording technology instead of shingled and which might be a 7200RPM drive at worst (for the purposes of this thread - I'm happy either way). Seemed pretty quiet when it was next to me in its external case and it's across the basement from me and I don't hear it now either installed in my HTPC/server. I got it for $175 off amazon which surprisingly was the best $/TB ratio compared to its lower capacity siblings as well.

Anyway, sounds like Chrispy_ already found his new drives but I thought I would share in case any gerbils are looking for cheap high capacity storage and don't mind giving seagate a chance. Final tangent - I've been using seagates for the last 4 years or so after I had a number of WD drives fail and the pricing of seagates has been irresistibly good.
i7-6700K | Z170X-Ultra Gaming | 32GB DDR4-2400 | EVGA GTX 1080 | 256GB SAMSUNG SM951 M.2 | 27" ACHIEVA 2560x1440 LED | Nanoxia Deep Silence 3 | EVGA SuperNOVA 550G2 | Win10 Pro x64.

Who is online

Users browsing this forum: No registered users and 1 guest