Why are NAS drives more expensive than desktop drives?

[Easy Rhino]

I was doing some research on new HDD for a NAS replacement. I noticed that an 8 TB "NAS" drive was $50 more expensive than its "compute" counterpart. Sure, the NAS branded drive is 7200 RPM versus the compute's 5400 but that seriously will not make a difference in typical home media workloads. These drives are not enterprise. Am I missing something?

 

[kayjay010101]

Probably better QA and therefore more expensive to produce.

That or it's just marketing. Probably the latter.

I typically just shuck drives for my NAS, but with the advent of SMR drives in the <8TB sector, I'd recommend that only if you're going above 8TB.

 

[CrAsHnBuRnXp]

NAS hard drives are rated for 24/7 use where as typical hdd's are not. they have tech to make them last longer to be more reliable.

 

[windwhirl]

NAS hard drives are rated for 24/7 use where as typical hdd's are not. they have tech to make them last longer to be more reliable.

On that, I think Seagate includes vibration sensors in their IronWolf HDDs, which are NAS products, while the Barracuda line doesn't have those.

 

[bmacsys]

The real reason? They can charge more money for them. That is how capitalism works.

 

[Bill_Bright]

I think it would be helpful, and more applicable if the question was not so general and instead was about two specific drives. At this point, we don't even know if the NAS drive you were researching is even made by the same manufacturer as the "compute counterpart".

Do the drives have the exact same seek times, data transfer rates, error correction? What about the type and size buffers? Are the warranties the same?

And yes, rotation speed does matter. 7200RPM drives tend to run hotter - if nothing else. NAS drives are often mounted in small enclosures that may have limited cooling compared to desktop drives that are mounted in cavernous (in comparison) cases.

The real reason? They can charge more money for them. That is how capitalism works.

Ummm, no. Let's leave politics out of this. The facts are, when NAS drives first hit the market, they did tend to cost more. But not today. With everything else being equal, the costs are typically nearly equal, with the NAS drives often costing less! Note NAS Hard Drives Vs Regular Hard Drives – Comparison where it says,

What Costs More? Normal Hard Drives or NAS Hard Drives?
An interesting myth about NAS hard drives and Ordinary Hard drives is about the price. Namely that NAS hard drives are a con to make us spend more money. Well, good news. That is NONSENSE! The price difference between NAS and ordinary Hard drives is less than 1-2% in most cases and (rather surprisingly) not always in the favour of NAS drives.

The myth that NAS drives are more expensive is based on the reaction at the time of release when the development of NAS hard drives was still comparatively experimental.

 

[rtwjunkie]

I was doing some research on new HDD for a NAS replacement. I noticed that an 8 TB "NAS" drive was $50 more expensive than its "compute" counterpart. Sure, the NAS branded drive is 7200 RPM versus the compute's 5400 but that seriously will not make a difference in typical home media workloads. These drives are not enterprise. Am I missing something?

Many NAS drives are also 5400 RPM also.

Normally the NAS drives have more vibration protection built into them, to reduce wear and tear in large volume cases. They may also have some higher operating temperature limits. I’ve got a couple of the oldest in my media server that have been running 24/7 for 8 years now.

 

[newtekie1]

Many NAS drives are also 5400 RPM also.

Normally the NAS drives have more vibration protection built into them, to reduce wear and tear in large volume cases. They may also have some higher operating temperature limits. I’ve got a couple of the oldest in my media server that have been running 24/7 for 8 years now.

They also support TLER. Which makes them far more compatible and reliable when used in RAID arrays.

The other thing is the Compute drives from Seagate can be SMR(256MB Cache on a 7200RPM drive is a good indicator of a SMR drive). Which for most probably won't be an issue, but if you copy large amounts of data to the drive, forget it. Once the CMR cache runs out, you're looking at write speeds down as low as 10MB/s. It can be painful.

 

[John Naylor]

Many reasons ...

1. Performance costs ... 7200 rpm has better performance
2. Warranty Costs - In many cases two HDs may be physically identical... warranty costs are essentially an insurance policy... the price premium ays for those RMAs diring the later years of the longer warranty.
3. Economies of scale - NAS drives are produced in lower quantities, the soft costs associated with production and inventory management are fixes ... when divided by 1 million units versus 5 million units, the large quantity has lower unit costs.
4. Yes, economics ... why do they charge more ? ... because they can. There are costs differences as stated above .... but the cost premium associated with that difference is not the only thing reflected in final costs. When you buy a sports package say on a new car that includes sportier rims and tires .... is the cost of that package = to the difference in costs between the two sets od rims / tired ? No never. Some times its even more than if you went out and bought them yourself.... and that way you have a spare set of wheels and tires. So yes, the reason the cost more is simply because people have shown a willingness to pay more, ... essentially the fundamental basis of capitalism. That's not a knock on the economic system, it's the cornerstone of our economy, 'what the market will bear".
5. And for an NAS think more about the usage than the enclosure. Few would suggest NAS drives for two drives installed in a desktop dedicated to network storage in a SOHO (Small Office / HOme) environment. Now if I pick up an NAS to serve this function do I suddenly need to buy NAS drives ? Of course not, the usage has not changed. If your network access is 24/7 with high I/O among users qwell into double digits doing simultaneous access, then I'd inverst in NAS drives.

 

[LFaWolf]

They also support TLER. Which makes them far more compatible and reliable when used in RAID arrays.

Just a quick note that TLER is a Western Digital terminology. Seagate has its own and it is ERC. Just in case you look for a NAS drive spec and don't see TLER. However, NAS makers such as Synology or QNAP don't use this thing in the consumer NAS enclosures.

 

[dragontamer5788]

And yes, rotation speed does matter. 7200RPM drives tend to run hotter - if nothing else. NAS drives are often mounted in small enclosures that may have limited cooling compared to desktop drives that are mounted in cavernous (in comparison) cases.

Note this Reddit Discussion

The direct way to measure rotational speed is via the acoustic frequency profile. If you have a disk spinning at 7200RPM (7200/minute = 120/second = 120Hz) then resulting vibrations will be at this base frequency or integer multiples (overtones) of it (120Hz, 240Hz, 360Hz, ...). For 5400RPM this would be multiples of 90Hz instead (180Hz, 270Hz, 360Hz, ...)

TL;DR: Dude held up acoustic frequency profiler, found that WD's "5400 RPM" drives made a noise similar to 7200 RPM, and therefore are most likely spinning at 7200 RPM. This includes some Reds (aka: WD's "NAS" line of drives). So be careful.

Imgur gallery for proof:

https://imgur.com/a/CzovEOQ

 

[rtwjunkie]

"5400 RPM" drives made a noise similar to 7200 RPM, and therefore are most likely spinning at 7200 RPM. This includes some Reds (aka: WD's "NAS" line of drives). So be careful.

Technically their Red NAS drives are variable speed, so it can be true sometimes.

 

[kayjay010101]

Technically their Red NAS drives are variable speed, so it can be true sometimes.

The problem lies more within the fact that they advertise "5400RPM-class", make them spin at 7200RPM, increasing heat, while the throughput is still that of a 5400RPM drive. So all of the negatives of slower drives without the positives.

 

[phill]

I was doing some research on new HDD for a NAS replacement. I noticed that an 8 TB "NAS" drive was $50 more expensive than its "compute" counterpart. Sure, the NAS branded drive is 7200 RPM versus the compute's 5400 but that seriously will not make a difference in typical home media workloads. These drives are not enterprise. Am I missing something?

I'm in the same boat, what I can say is that I'm not going for a NAS drive and I'm going to go with a Enterprise drive instead. The warranties are better, they are rated for a higher workload that I'll never hit and they'll probably be faster.. I'm going to see what happens and go from there

Please let me know what you go for yourself as I'd be interested in the route you take

I'm looking at the Seagate Exos models I'm not quite sure yet what size, but 8TB are a minimum since I'd already got 4TB Reds in my Synology system...

 

[Easy Rhino]

I'm looking at the Seagate Exos models I'm not quite sure yet what size, but 8TB are a minimum since I'd already got 4TB Reds in my Synology system...

I also have a pair of 4 TB drives that need to be replaced. They are about 6 years old and nearly full. I want to replace with 8 TB drives.

 

[Solaris17]

FWIW, I run 4x 8TB SG exos drives in my QNAP and I push a lot of data through it. works really well.

 

[Bill_Bright]

Technically their Red NAS drives are variable speed, so it can be true sometimes.

Right. And that's a good thing.

It seems some folks are complaining because these drives offering better performance than their marketing specs. These drive never stop spinning unless powered off completely. That's a good thing. Instead, they slow down when idle and ramp up with being accessed. Again, a good thing.

Is the Intel Core i7-9700K criticized because it is not marketed as a 4.9GHz processor - even though it has that speed for its turbo mode? No.

Do these drives fail to meet published specs? NO!

The problem lies more within the fact that they advertise "5400RPM-class", make them spin at 7200RPM, increasing heat, while the throughput is still that of a 5400RPM drive. So all of the negatives of slower drives without the positives.

What? NO!!! That is the wrong conclusion! And where does it say the throughput is the same as 5400RPM? Answer: NOWHERE!

What I am reading is these drives spin at 7200RPM but consume the power and generate the heat of 5400RPM. How is that bad?

 

[dragontamer5788]

Technically their Red NAS drives are variable speed, so it can be true sometimes.

I don't believe so. If the NAS were variable speed, we'd see both the 90 Hz (90 rotations/second, or 5400 RPM) and 120 Hz (120 rotations/second, or 7200 RPM) noise levels.

But look at the noise levels: the drive spins up to a constant 120 Hz (7200 RPM) and stays there. There's no other noise, indicating that the read/write mechanisms are inactive before the 120 Hz mark.


----------

This isn't "variable speed" behavior. This is "accelerate to 120 Hz" behavior, with no indication at all that 5400 RPM is even supported.

What I am reading is these drives spin at 7200RPM but consume the power and generate the heat of 5400RPM. How is that bad?

Spinning faster uses more electricity, its that simple. These drives are spinning at 7200 RPM, and therefore generating the power and heat of a 7200 RPM drive.

There are certainly lower-power 5400 RPM drives out there, but it isn't a WD Element 8TB (WD80EMAZ-00WJTA0), despite the marketing. WD is really screwing their technical specifications recently, we can't trust them on SMR or 5400 RPM indicators.

EDIT: Here's some proof (from: https://arstechnica.com/gadgets/2020/09/western-digital-is-trying-to-redefine-the-word-rpm/)

The Seagate 5400 RPM has 5.3W of power (operating) and 3.4W of idle power.

The WD Red uses 8.8W of power and 5.3W at idle: drawing 60% more power than a "true" 5400 RPM drive.

 

[silentbogo]

I was doing some research on new HDD for a NAS replacement. I noticed that an 8 TB "NAS" drive was $50 more expensive than its "compute" counterpart.

That's because after recent controversy all NAS-marketed drives are CMR. Before that the difference was negligible for the same spec (before SMR they were pretty much identical, just balancing out perf vs reliability for different versions).

The other thing is the Compute drives from Seagate can be SMR(256MB Cache on a 7200RPM drive is a good indicator of a SMR drive). Which for most probably won't be an issue, but if you copy large amounts of data to the drive, forget it. Once the CMR cache runs out, you're looking at write speeds down as low as 10MB/s. It can be painful.

That's if you get lucky. Had a pair of 4TB Compute drives (unfortunately bought just a month before news hit the media), and after less than 1 year of use in RAID-1(md) one drive became super-loud, to the point where it was louder than my shitty washing machine. The second one works as a standalone drive now, but the speed drop is noticeable, especially when transfering 100+GB of worklogs or backups over LAN. Sync times also became slower for each consecutive run, even if disks were unchanged between runs.

Spinning faster uses more electricity, its that simple. These drives are spinning at 7200 RPM, and therefore generating the power and heat of a 7200 RPM drive.

It's not about power usage. Faster RPM means more wear on mechanical parts and higher chance of long-term consequences.

 

[Bill_Bright]

Spinning faster uses more electricity, its that simple.

No! That is a blanket statement and like all blanket statements, it is wrong (and yes, I see and understand the irony of my own blanket statement).

EDIT: Here's some proof

And that's not proof of anything!

You are assuming all 8TB drives, regardless of brand, are 100% mechanically and electronically identical (other than rotation speed, in your example). That is wrong!​

You are assuming all drive motors have the same electrical efficiency. Wrong!​

You are assuming all drive motors have the same friction coefficient. Wrong!​

You are assuming all drives have the same number of platters. Wrong!​

You are assuming all drives have the same cooling efficiency. Wrong!​

You are assuming all the motor's and drive's moving parts (platters, spindles, armatures, etc.) have the same total weight and mass. Wrong! Wrong! Wrong!​

If you have two 100% identical drives, then yes, it takes more power to spin the platters up to, and to maintain the faster speeds. And thus, more heat due to friction and electrical efficiencies will occur too. But it sure is NOT "that simple" when it comes to "all" drives - especially when the drives are NOT 100% identical.

The fact that Seagate drive and that WD drive have significantly different standby/sleep specs (.25w vs .8w) clearly indicates those two drives are far from 100% identical.

 

[dragontamer5788]

And that's not proof of anything!

90 Hz tone indicates a 5400 RPM drive.

The Seagate is actually 5400 RPM, as advertised. Not only does this lower the Hz tone from 120 (aka 7200) to 90 (aka 5400), this also comes with improved heat and noise characteristics.

The WD Reds are 7200 RPM. Which is fine, I'm actually a fan of 7200 RPM, but what I'm against is WD mislabeling their hard drives on a consistent basis.

 

[Bill_Bright]

If you think it is misleading, don't buy them.

 

[dragontamer5788]

If you think it is misleading, don't buy them.

I don't buy WD anymore, because of this. Toshiba is my goto hard drive manufacturer right now.

I presume other posters around here would be interested in the information. So I'm just making sure others know of the misleading marketing that WD is operating under. Be careful, WD says "5400 class" but doesn't really mean "5400 RPM". They're not technically lying, but its highly misleading.

Otherwise, you'll be surprised by the +50% power and heat that comes from your WD "5400 class" drive (which actually eats up power and heat like a 7200 drive)

 

[Bill_Bright]

your WD "5400 class" drive

which actually eats up power and heat like a 7200 drive

But you are making assumptions again. Yes, if you compare that specific Seagate and the WD you hand picked earlier, your claim is true. But you cannot, and must not assume that applies to all drives.

And they are not my WD drives. In the last 5 years, I've gone with Samsung SSDs exclusively in all our builds - even in my NAS.

And just for the record, do a search on me and "marketing weenies" on this site and you will see I have had a long war against the marketing tactics (and many executive decisions) of many companies. That does not mean the products they sell, or the designers, programmers or hardware technicians who make them, are spewing the "hype" or are being deceptive, or are making lousy products.

IMO, it is important to separate the marketing hype from the product and judge each independently. We must not let our biases and disdain for the company (or their policies) influence our opinions about the products they sell.

I mean, are you in the market for a new truck? If you listen to the hype, Ford, Chevy, GMC, RAM, Toyota and Nissan all make the #1 truck - depending on which criteria to decide ranks highest.

Want the #1 browser? Flip a coin - they all are #1 in some category or another.

 

[dragontamer5788]

And they are not my WD drives.

I can only assume that the generic you is not used very much in your dialect of English? American English and the way we talk is different from state-to-state. But in the part of the country I come from, what I was doing there is called a generic you. I'm not suggesting that you own WD drives, I'm suggesting that a "generic you" would be unhappy to be surprised by a 7200 RPM drive, after the "generic you" bought a 5400 RPM drive.

Actually, its usually called an "indefinite you" around here, but I guess Wikipedia calls it "generic you".

But you are making assumptions again. Yes, if you compare that specific Seagate and the WD you hand picked earlier, your claim is true. But you cannot, and must not assume that applies to all drives.

I do not have the time to look for audio-tests for 120Hz or 90Hz signals every time a new hard drive comes out.

WD's marketing is unclear. See it for yourself: https://documents.westerndigital.co...uct-brief-western-digital-wd-red-plus-hdd.pdf

The WD80EFAX is well known to be a 7200 RPM drive through acoustic tests. WD has mislabeled their drives and refuses to update their documentation. This is still happening today (9/30/2020), I just double-checked their spec-sheets.

If I can't trust WD's spec sheets, then what should I trust? As I stated before: I actually prefer 7200 RPM drives, but more importantly, I demand accurate documentation. WD has shown that they are no longer accurately documenting their drives.