If you use Backblaze as your reference, you're ill informed. Backblaze started their business after flooding took out several HD factories (Thailand if I recall correctly) and jumped in to the server market by using consumer drives in a server environment. Many concluded, wrongly, that this was just the equivalent of a stress test, like using P95 on a CPU. That assumption simply couldn't be more wrong. It is, in fact, the protection schemes in consumer drives that actually cause them to fail in a server environment.
but before getting into that let's look at the BB business model:
In short, by its own admission, Backblaze employed consumer-class drives in a high-volume enterprise-class environment that far exceeded the warranty conditions of the HDDs. Backblaze installed consumer drives into a number of revisions of its own internally developed chassis, many of which utilized a rubber band to "reduce the vibration" of a vertically mounted HDD.... a heavy HDD is mounted vertically on top of a thin multiplexer PCB. The SATA connectors are bearing the full weight of the drive, and factoring the vibration of a normal HDD into the non-supported equation creates the almost perfect recipe for device failure.
The Backblaze environment employed more drives per chassis and featured much heavier workloads (both of which accelerate failure rates tremendously) than the vendors designed the client-class HDDs for. This ultimately helped Backblaze save money on their infrastructure. The Seagate 3 TB models failed at a higher rate than other drives during the Backblaze deployment, but in fairness, the Seagate drives were the only models that did not feature RV (Rotational Vibration) sensors that counteract excessive vibration in heavy usage models -- specifically because Seagate did not design the drives for that use case.
A server environment dictates thick concrete floors and structural sound rack systems. For that reason, there is no need to include rotational vibration sensors in server drives. Mounting drives vertically on a case laid on its side on a table and holding them in place with rubberbands ... it's as if they designed the environment on a whim saying "lets see how fast we can make a drive fail".
More importantly perhaps, consumer drives are designed with a feature called "head parking" ... when not spinning, the arm is 'parked" away from the platter so that when the copy paper delivery guy bumps ya desk with his hand truck or ya pet lab sleeping under ya desk jumps up when the UPS man knocks on the door, the hed doesn't crash into the patters causing damage. Consumer drives are rated for 250k - 500k parking cycles ... way more than the typoicl consuer drives ever experience cause of low I/O. OTOH, server drives they are all about I.O and they can burn through those cycles in a matter of months ... server drives do not come equipped with that feature because server farms do not have cases laying sideways on a desk and held in with rubberbands. This is why you don't use server drives on a desktop and you don't use consumer drives in a server farm
If you are using server drives on a desktop ...
no they are not more durable, far from it ... they have no built in protection.
Real world failure rate / RMA data was published for HDs every 6 months by the french site behardware. They stopped doing so, dunno why ... can't read french) But the data is in numbers
Period Ending | HGST | Seagate | Toshiba | Western |
2017-08-01 | 0.82% | 0.93% | 1.06% | 1.26% |
2016-12-09 | 1.13% | 0.72% | 0.80% | 1.04% |
2016-05-13 | 0.60% | 0.69% | 1.15% | 1.03% |
2015-11-09 | 0.81% | 0.60% | 0.96% | 0.90% |
2015-05-19 | 1.16% | 0.68% | 1.34% | 1.09% |
2014-11-06 | 1.01% | 0.69% | 1.29% | 0.93% |
2013-04-30 | 1.08% | 0.86% | 1.02% | 1.13% |
2013-10-30 | 1.16% | 0.95% | 1.54% | 1.19% |
2013-05-10 | 2.40% | 1.44% | 1.15% | 1.55% |
| 3.45% | 1.65% | NA | 1.44% |
Average | 1.36% | 0.92% | 1.15% | 1.16% |
So that's it, the real data based upon actual HDs replaced under warranty that were between 6 and 12 months of usage. Each report listed the data for the current 6 month period as well as the previous 6 month period. So what can we glean from this information.
1. This is real data from consumer drives installed in a consumer environment.
2. Fanbois may wanna beat their chests if they like the numbers for 'their brand", but what is shows me is that out of each 1,000 drives ... the guy with the best number delivered 991 good drives ... the guy with the worst number delivered 986 good ones... that's pretty damn close.
3. Failure rates were pretty consistent for the last 4 years of the data set.
It must be noted that drive models rather than brands had significant variation. Some models had failure rates as high as 10% ... some were under 0.50 %
As to which HD to buy ....
1. Pick the drive appropriate for your usage.
-Use NAS drives in NAS devices
-Use consumer drives in desktops
-Use Green drives if you have 100's of drives and energy is important to you. There is nothing else of value here and they have significantly slower performance (5400 rpm)
-Use server drives in server farms ... no server drives are not tougher, they are missing important desktop features like head parking
-Use surveillance drives in surveillance applications
2. Check the warranty ... be aware that often the same drive is sold under different model numbers... a drive with a 2 yeat warranty for $55 ... may be the same drive as the 3 year warranty model that sells for $67. We only buy models with 5 year warrantees.... the cost difference is minimal ... it's not the 'free replacement, it's the T & E replacing the drive that matters.
3. Consider not buying a HD. In THGs testing, the Seagate 2 TB SSD was 1.54 times faster in gaming than the WD Black ... 2.5 times faster than the WD Blue*. We haven't puchased a HD in about 8 years. Have about 100 SSHDs installed over that period and no reported failures. I have about a dozen here, 4 of which are > 5 years old.
For comparison purposes:
WD Black WD2003FZEX = 2 TB / 5 year warranty / 0.45% failure rate / 100% relative gaming performance / $105 on newegg
Seagate Desktop SSHD ST2000DX001 = 2 TB 5 year warranty / 0.43% failure rate / 154% relative gaming performance / $885 on newegg
