Hard Drive Reliability Data

Apr 3, 2010
16,022
0
0
Seattle, WA
#1
Companies that offer cloud storage can be a great source of reliability for hard drives as they go through thousands of them. This is one such source: https://www.backblaze.com/blog/best-hard-drive/

"Over the past year we’ve been releasing hard drive reliability statistics based on the drives we use to store customer data for our online backup business. As of the end of Q1 2015 we had 44,252 hard drives spinning in our datacenter. If we subtract boot drives and drive models with less than 45 drives from that total, we get 42,749 hard drives remaining spread across 21 drive models. Below are the hard drive reliability statistics for these drives for Q1 2015."

Click on the link to see the individual stats for each hard drive model. Most have low single digit failures with the exception of these (the column before last is the percentage failure):

Code:
Seagate Barracuda 7200.11
(ST31500341AS)	1.5TB	259	5.0	31.68%	20.1% – 47.5%

Seagate Barracuda LP
(ST31500541AS)	1.5TB	1,485	5.1	12.16%	8.9% – 16.3%

Seagate Barracuda 7200.14
(ST3000DM001)	3TB	485	2.4	26.65%	19.6% – 35.4%

Seagate Barracuda XT
(ST4000DX000)	4TB	175	1.9	4.62%	0.6% – 16.7%

Western Digital Red 3 TB
(WDC WD30EFRX)	3TB	1045	0.9	12.87%	8.5% – 18.7%

Western Digital 4 TB
(WDC WD40EFRX)	4TB	45	1.0	9.01%	0.2% – 50.2%

Western Digital Red 6 TB
(WDC WD60EFRX)	6TB	450	0.3	7.77%	3.4% – 15.3%
For audio use, there is no reason to buy these high-performance drives. Get the green version of Western Digital for example. As for Seagate, boy this is shameful scores.
 

ack

VIP/Donor & WBF Founding Member
May 6, 2010
5,200
20
38
Boston, MA
#2
I would come from a slightly different angle... First of all, Seagates are indeed truly terrible drives - an external Seagate USB drive for backups I was using, died last week with very few hours on it (coincidentally, right after a full back-up); for a variety of reasons, that set me off to build my next PC (I always build them, haven't bought one for over 25 years)... this time, I went all-in, sorta, with reliability as top priority. To that extent, you can't beat SSDs, a) for speed, b) even when they go bad, they become read-only and you don't lose data. So Samsung has remarkable SSDs (as do others, in fact), and I picked an 850 Pro; for mechanical, I picked a 4TB HGST 7K4000 Ultrastar with 2,000,000 hours life expectancy (in layman's terms); there are also the 7K6000 series, all the way up to Helium-cooled 10TB drives. I do a lot of video productions and transcoding speed and hardware reliability [now] means a lot to me (having suffered through a number of failures). If anyone cares, I also got the i7-5820K unlocked 6-core CPU with 2400 DDR4 memory, and overclock both (the CPU at 4.0GHz, the memory at 2400 - "stock" speed for that CPU is 2333), coupled with a very good GeForce... blazingly fast and reliable, and dead quiet (and I mean that) with Noctua fans; I just finished all stress-tests, and they all passed.

So I would argue for hardware reliability, especially drives, over anything else. With storage prices being so low, why not get a WD Red, or equivalent.
 
May 19, 2014
714
3
18
Round Rock, TX
#3
Companies that offer cloud storage can be a great source of reliability for hard drives as they go through thousands of them. This is one such source: https://www.backblaze.com/blog/best-hard-drive/

"Over the past year we’ve been releasing hard drive reliability statistics based on the drives we use to store customer data for our online backup business. As of the end of Q1 2015 we had 44,252 hard drives spinning in our datacenter. If we subtract boot drives and drive models with less than 45 drives from that total, we get 42,749 hard drives remaining spread across 21 drive models. Below are the hard drive reliability statistics for these drives for Q1 2015."

Click on the link to see the individual stats for each hard drive model. Most have low single digit failures with the exception of these (the column before last is the percentage failure):

Code:
Seagate Barracuda 7200.11
(ST31500341AS)	1.5TB	259	5.0	31.68%	20.1% – 47.5%

Seagate Barracuda LP
(ST31500541AS)	1.5TB	1,485	5.1	12.16%	8.9% – 16.3%

Seagate Barracuda 7200.14
(ST3000DM001)	3TB	485	2.4	26.65%	19.6% – 35.4%

Seagate Barracuda XT
(ST4000DX000)	4TB	175	1.9	4.62%	0.6% – 16.7%

Western Digital Red 3 TB
(WDC WD30EFRX)	3TB	1045	0.9	12.87%	8.5% – 18.7%

Western Digital 4 TB
(WDC WD40EFRX)	4TB	45	1.0	9.01%	0.2% – 50.2%

Western Digital Red 6 TB
(WDC WD60EFRX)	6TB	450	0.3	7.77%	3.4% – 15.3%
For audio use, there is no reason to buy these high-performance drives. Get the green version of Western Digital for example. As for Seagate, boy this is shameful scores.
The power scheme of your system could have a significant effect on HDD longevity. There is a direct relationship between the power savings level and HDD ramp up/down times. More power savings = decreased HDD longevity. I didn't see any links so I'm assuming this wasn't accounted for in the tests since you said boot HDDs were excluded.
 
Apr 3, 2010
16,022
0
0
Seattle, WA
#4
The power scheme of your system could have a significant effect on HDD longevity. There is a direct relationship between the power savings level and HDD ramp up/down times. More power savings = decreased HDD longevity. I didn't see any links so I'm assuming this wasn't accounted for in the tests since you said boot HDDs were excluded.
These are not tests. They are actual failure rates of drives used in that company's online cloud storage service.
 
May 19, 2014
714
3
18
Round Rock, TX
#5
These are not tests. They are actual failure rates of drives used in that company's online cloud storage service.
ok, thanks Amir. I assume there is no data WRT the configuration of systems attributing to such failure data?
 
May 19, 2014
714
3
18
Round Rock, TX
#7
OK, just read the complete thread. Here's the problem I have with the data. a) no configuration information is provided. Are all the drives configured with the same OS, apps, etc? b) Are all the drives subject to the same power up/down times from both a frequency and time (# of minutes up/down)? c) Do all HDDs use the same power scheme? d) What were the HVAC conditions, all the same? e) Any situations with power surges/brownouts/blackouts? f) Were all HDDs brand new or refurbished? I assume all are new? g) What conditions were the HDDs subject to? (reads, writes, sequential, random) and how often?

Net - Taking my questions to an extreme - HDD A:Situated in a rack closest to HVAC (coolest temp, lowest humidity) experienced no power anomalies, experienced hourly reads, writes both sequential and random but not for extended time intervals (e.g.: max reads/writes for 10 hours consecutively), was brand new and was never shut down from the system level or from any power scheme vs HDD B: 180 degree opposite conditions of HDD A which would last considerably less days. There is not enough information to make an accurate assessment IMO.
 
Apr 3, 2010
16,022
0
0
Seattle, WA
#8
They have a datacenter with 44,000 drives. It is to their benefit to keep all the equipment in proper temp as to reduce their costs. That drives fail, is not something they want to see.

But yes, it is a sample data point and should be taken for whatever value one puts on it.
 

Mosin

[Industry Expert]
Mar 11, 2012
894
0
0
#9
Samsung offers a ten year warranty on their pro ssd, so that alone is enough to make conventional drives obsolete, as far as I'm concerned. I'm in the planning stages for a new computer build, and it will be 100% solid state.

Besides, you can't build a triode computer. :D
 
May 19, 2014
714
3
18
Round Rock, TX
#10
They have a datacenter with 44,000 drives. It is to their benefit to keep all the equipment in proper temp as to reduce their costs. That drives fail, is not something they want to see.

But yes, it is a sample data point and should be taken for whatever value one puts on it.
Agree Amir, but location of racks and maintenance (dust is the enemy of computers) has a significant factor on longevity. Someday we'll be all SSD and this won't matter nearly as much....
 

rbbert

Active Member
Dec 12, 2010
3,423
0
36
Reno, NV
#11
One problem with the potential usefulness of this data is that none of the drives I typically see on sale for home use are on this list, and as has been noted just because some of a particular manufacturer's drives do poorly doesn't mean all will. For example, some of the Seagates do very well.
 

About us

  • Founded in 2010 What's Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals we enjoy learning about new things and meeting new people and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing