Using a 16TB WD Gold mechanical hard drive with TLER on Desktop without RAID - downsides?

[c627627]

Western Digital weekly sale has 16TB Red Pro and WD Gold both on sale for $299.
Paypal store had 12% cashback so effectively $263 for WD Gold 16TB - I pulled the trigger.

First, it's not really clear why I shouldn't choose Gold over Red Pro, since Gold Specs are better, minus energy consumption, which, what would that be in real dollars, savings on energy between the two?

Second, basic search says do not use WD Gold in desktops, it is overkill for its expense, plus TLER is not good for Desktop non-Raid.

It's always a problem when they start talking about money because for some reason, it never occurs to them that eventually, like this week, these things DO go on sale. And they become cheaper than regular stuff, so money is no longer the reason, TLER is. Here are two quotes I pulled from the Internet.

Post your thoughts on the downsides of using TLER on a regular desktop, [all things being equal with pricing],
was it a good or a bad idea to go with WD Gold?


Quotes:
"WD Gold is ... a higher quality drive than WD Black - but it is not as suitable for desktop use.... since the Gold... drives have TLER-Time Limited Error Recovery - which limits the amount of time a drive will attempt to recover an error. TLER is necessary for drives used in RAID to reduce the chance a drive will drop out of the RAID due to excessive time being taken to recover form an error. Desktop drives, like the WD Blacks, have no need for TLER and thus are less prone to losing data due to errors..."

and

"Drives with TLER are specifically designed for redundant RAID arrays. TLER is a drive function, where if the drive has difficulty accessing a portion of data, it will give up quickly and report an unreadable condition to the host controller. This timing is usually around 6-7 seconds maximum. The reasoning for this is that in a redundant array, all the data can be accessed or reconstructed from parity using other drives. The host controller then uses the remap function on the hard drive to mark those sectors bad, and writes the reconstructed data to spare sectors. This whole process happens seamlessly and without interruption or degradation of the RAID array. If it happens so many times that it runs out of spare sectors, the drive will be dropped from the array and marked as bad.

If you use a drive configured with TLER in a non-redundant configuration, such as a single drive or in a RAID0, the drive still acts the same - it quickly gives up reading the data. This will usually cause CRC or other errors to display in the operating system, and the data will not be able to be read. Naturally, this results in a loss of that data.

In contrast, a drive designed for use in desktop systems as single drives don’t have this timeout. Without TLER, the drive will literally try forever to get that data (and more often than not will eventually succeed). The operating system will appear to be extremely slow or frozen at this time, as it is waiting for the drive to become responsive again. If a drive without TLER is used in a redundant RAID, it will essentially cause the RAID to be degraded immediately upon hitting a single unreadable sector since the drive will appear to have become unresponsive to the host controller.

The short answer of it is…
To avoid unnecessary down time and headaches, don’t use a drive without TLER in a redundant RAID array.
It’s OK to use a drive with TLER in a non-redundant configuration, just be aware it will be more difficult to recover data from sectors that develop issues."

 

[don256us]

If you are just using a single drive, then the TLER would be a deal breaker. However if it were part of an array, it's a deal maker.

 

[c627627]

Just a single drive with one 16TB partition.
So at first it seemed like a deal breaker but the more I read the more foggy it gets...

So let's talk about it, there are no real 16TB alternatives, so I got this 16TB drive, what's going to happen?

 

[don256us]

If you already have it, enjoy it. This feature may actually prove to alert you about data loss early as it ages out and begins to fail.

 

[c627627]

Well it's on its way, so I don't have to use it yet, still time to talk about it.

So you start out believing that TLER is some sort of a feature that makes you stay away from these drives if you have a Desktop PC, period.
Then reality sets in and the more you read - the more you realize it's not as bad of an idea as it seems at first...

Kind of looking for a devil's advocate conversation, so first of all it's 16TB, considerably more than what's offered in other series, next... WD Gold is supposed to be THE most reliable drive Western Digital has, which, wait a minute, who wants to lose 16TB of personal data, this would be for essentially everything I have to store. I mean, there will be a backup spread across smaller drives but this thing is going into my PC for all my storage data.

So what's a better solution? I mean I asked my self the same question and if I were to go for a (much) smaller WD Black, first of all cost per GB is A LOT higher, second of all, I ask myself why.. so that what... my data will be safer.... than... what? On WD Balck than on WD Gold, which is THE most reliable hard drive WD has in its arsenal!?

There is a great deal of confusion on the net about this, because one of the arguments against WD Gold is money - but this argument is so short-sighted (!) because time passes and 'prohibitively expensive' goes on sale for less than other drives are per GB.... Anyway, just looking for some educational discussion on why go with option A . vs option B. in 2022?

 

[EarthDog]

I'm not entirely sure what (if any) problems it causes using a drive with TLER while not in a RAID array. I'm a simpleton and use the drives as what they are marketed for, lol. Forced to guess from what i know/read, I do not believe there is any harm using one alone.

As far as reliability, the difference between most models is nearly negligible (just looking at backblaze data.. not the best source, indeed, but better than most other datasets). Theres always some potatoes that tend to fail early on every line. I wouldnt necessarily hang my hat on there being a worthwhile difference (not enough to capitalize 'THE' twice )... unless you don't have a duplicate copy of your data which is more a function of a poor backup plan than anything.

Me.... I run two WD Green drives for backups. One is in the PC, the other a copy is offsite (at my moms, lol... was in a drawer). Every month or so I'll sync them up manually and this has worked well over the past decade.

RE: The Black drives and price... that makes sense to me. The blacks are designed for performance and is their top of the line product for desktop performance/gaming use. The Gold is simply a drive with TLER which helps for RAID ecosystems (all drive makers have some form of this, note). I dont know why you'd pay a premium for a performance drive and use it for backups. But I'd be plenty happy with a new green or blue for backups too. It's cold storage and the drives don't get beat on either.

 

[don256us]

For data retention, I prefer multiple smaller drives with a parity backup. If I lose a drive, my data is just waiting for a replacement drive and a rebuild.

The TLER feature seems like it might restrict data flow when the drive is struggling to read a sector. A non-TLER drive will try longer to read data thus giving you a better chance to save data prior to complete loss.

If given the choice and the price is the same, I'd purchase based on it's advertised purpose.

 

[c627627]

So let's than shift from theory and general talk and say what are actual alternatives to WD Gold, comparing them to WD Gold's 5-Year Warranty and 2.5M hours MTBF mean time before failure [does MTBF figure mean anything]? WD Gold is advertised to handle 550TB per year. I mean so what if you don't need it that much usesage, isn't it an indication of overall longevity/reliability?

The argument on price should include the fact that the price was for 16TB WD Gold directly from WD itself for $299 - 12% Paypal cashback = $263 for 16TB WD Gold.


And PS: TLER is generally used out there on the internet as an argument against WD Gold in a home non-raid PC... The other one being the price, which completely ignored the possibility that WD will ever have a sale on WD Gold, which they did, so I think we're just left with the TLER argument, which I don't think was ever adequately explained as far as what the overall situation is as far as comparisons of actual reliability... leaving people who don't quite understand technical trerms scratching their heads about what to do when it comes to $263 for 16TB of WD Gold, adding that many have additional senior or edu discounts when buying from WD, so that $263 could even be 10%-15% less...

 

[EarthDog]

It means it should handle that amount of writes. That doesn't mean it won't fail more/less in other ways.

As far as competition goes, Seagate also makes some solid drives with similar technology (they call it Error Recovery Control, ERC). Hitachi/Toshiba do as well (called Command Completion Time Limit, CCTL). I don't know the pricing/warranty/TB per year info offhand, but, you can easily take a look to see what's up.

I think you should get the Gold for a great price and enjoy as TLER really doesn't seem to be a problem for a single drive.

 

[c627627]

Okay got it on number of writes.
And what of mean time between failures - have MTBF figures meant anything historically vs. real time failure figures?

 

[don256us]

TLER is not a deal breaker for me nor do I have a strong argument for its use as a lone drive. It's just that if TLER restricts the time the drive will take to read data and the benefit of this is only found in a RAID configuration, I wouldn't want it. Now when new and when working in perfect order, I do NOT see an issue with TLER. Therefore my argument doesn't even kick in until the drive starts to fail. It is only in its last days do I see the TLER as a potential issue to data recovery. Note that I am using vague terms here because its not like I know how it will behave in 3-7 years.

However, if you find other benefits that out-weigh TLER in standalone, then its worth it. I just wouldn't buy it based on your need and its design. Especially since you found the price of the RED to be the same which means that my other consideration was the same. With all things being the same and your usage, I'd have gone with the RED but my argument against the Gold in this case doesn't even kick in until its end of life so consider that.

 

[EarthDog]

It's just that if TLER restricts the time the drive will take to read data and the benefit of this is only found in a RAID configuration, I wouldn't want it.

But it doesn't do that... right? Only on bad sectors as I read it?

 

[don256us]

But it doesn't do that... right? Only on bad sectors as I read it?

But that's my point. It's fine as long as everything is working great. At the end of life when sectors start to go bad, that's when I see the issue. Not before that.

 

[EarthDog]

Personally, I see that as a good thing... a method to prevent bad sectors from affecting the drive (sort of). Also it's another 'symptom' to see if the drive is going bad. Not like HDDs fail in the same way as SSDs (tend to simply poof/done), but, at least it's another symptom before failure.

I mean, when your drive has issues (bad sectors) you get a new one, right? How does this prevent/delay/change that?

 

[c627627]

You are asking great questions.
I think they are saying that TLER gives up on recovering data quicker, and non-TLER keeps hammering at it so it almost always eventually gets the data, after a longer while.

My thinking has to do is *when* the eventual failure will happen.
Gold has a MTBF figure that is much higher that any other drive.

2.5 Million hours for Gold and 'only' 1 Million hours for Red Pro.

So as a matter of a friendly discussion: why should we go for Red Pro if MTBF for Gold is 250% higher!?
Does Mean Time Before Failure figure mean anything in real life?


AND... the 16TB Red Pro, it has TLER too, doesn't it?
So why would anyone go for Red Pro if pricing is the same for Gold?


SPEC SHEETS:

 

[EarthDog]

Because.......... does it matter? 1M hours is 114 years if I mathed right, lol. I don't worry about MTBF in devices like this with values like we see. Like write longevity, unless I plan to use these well into my 140s, lol, MTBF isn't a worthwhile metric (to me).

But seriously... how does (or doesn't) having TLER change how you(royal you, drive owners) behave with a bad drive? I don't think it changes much at all... and if anything, is better for warning you (assuming I'm understanding the technology properly, of course).

the bottom line is that, no matter the metric, we have no idea when a drive will fail. You can hedge your bets and shop for the highest this and the lowest that, but in the end, the differences are negligible in most cases. A proper backup plan (read: redundancy) helps to mitigate failures....not the pittance of difference between hardware.

 

[c627627]

Your answer about MTBF is relevant, if true.
We have the same company, using the same test method, stating officially that one drive has a MTBF figure that is 250% higher than another of their drives.
I would love to ask a storage expert. Customers (royal us) ask, does that mean one drive will last 2.5 times longer (on average)?
Will it last 100% longer, or 50% longer?

How about this: will the drive with 2.5 times the MTBF of another drive last longer (on average), period?

I mean, friendly question: isn't it a pretty bold statement to completely dismiss the MTBF figure? But if you ever talk to a storage company rep at CIS, that's a crucial question to ask. If you guys go next January maybe you can ask them?


I don't look at it as 114 years, I look at it from the point of view of having a million Red Pros and averaging out the time before they all start to fail and comparing it to the average time a million Golds start to fail. Which is higher? How much higher?

As a matter of a friendly discussion: If that figure is in any way connected to MTBF figure, then what could possibly be the reason for us to choose Red Pro at the same price?



Answering your question about TLER, from what I have researched, they say that TLER gives up too soon so you lose your data. Whereas a non-TLER just keeps hammering and eventually recovers the data, in other words, so you DO NOT lose your data. That's what I understand them saying. But I say when your drive starts clicking for ten minutes, it's game over in real life when it comes to royal us continuing to use that drive.


Now to me, they don't see the big picture, which is if MTBF has any meaning, Red Pro will start to fail in year X. But Gold will start to fail in year 2.5X, or [some number] times X.
Now that approach would be totally shattered if what you say is true: That MTBF is *completely* meaningless.

Intuitively, my mind goes, okay maybe not 2.5 times like they say, maybe not 250% more, but surely some [significant] % percentage more.
It's a really good question to ask a storage expert if, or rather when, you have a chance.


And going back to another question, does 16TB Red Pro have TLER? If yes then what are we even talking about, and why choose Red Pro again?

 

[don256us]

The more I look at it the less difference I see. The Red Pro and the Gold could be the exact same product but they aren't. Since ED started down this road, for fun I found that the Gold's MTBF = 285.39 years. we all know it ain't gonna last that long.

I see that the Red Pro is CMR but I haven't seen if the Gold is CMR or SMR. I'd guess CMR until I see otherwise. CMR is the better choice if a lot of writes will be done. SMR is OK if it mostly reads with much fewer rights.

Both have the 5 year warranty. Both spin at 7200 RPM. Both have 512MB cache.
Red Pro = NAS
Gold = Enterprise

When looking at various sites comparing the two, little difference is found.

 

[c627627]

So does Red Pro 16TB have TLER?
Since speed and cache are the same, and you found little difference in performance, we are back to which will last longer.


What does 2.5 million hours vs 1.0 million hours MTBF tell you don? I mean if it's a gimmick, that's fine, I just don't know.

 

[don256us]

If the MTBF were a good indicator, I'd think that the warraty would reflect it. It doesn't. Both of these drives have a 5 year warranty.

Post magically merged: Mar 29, 2022

I have not seen the Red Pro list TLER specifically but it does have a lot of other letters to "prove" it's worth as a NAS drive.

Post magically merged: Mar 29, 2022

I have no clue about MTBF. It must have to be a mathematical calculation that a bit will lose or change polarity. I suspect that it's more of a marketing number as the number itself is so large. Perhaps it means something like: "this is how often a bit will get lost but the sector is fine and that part of the drive will still hold data it just happened to fail this one time" kind of thing? This would be different than the sector itself going bad and thus being marked not to be used in the future kind of thing.

I don't know. What I do know is that if it really meant something for average users, we'd see it reflected in something more tangible like the warranty.

Post magically merged: Mar 29, 2022

mtbf formula - Google Search