Western Digital gets sued for sneaking SMR disks into its NAS channel

[Delicieuxz]

2 minutes ago, leadeater said:

also that post is 100% relevant here 

Where there's the argument of 'If the change broke ZFS performance, then ZFS is the problem' going around, the designer comment of "WE DO NOT BREAK USERSPACE!" has relevance.

 

You can think of it what you will. It's a relevant perspective, though.

[Curufinwe_wins]

Nothing more really to say from this thread though honestly. 

 

[leadeater]

18 minutes ago, Curufinwe_wins said:

Hard to say that a 50% reduction in performance for moderate file copies is reasonable for what is marketed by spec as a pure performance improvement.

 

(Note I HATE watching videos when articles work just as well, I hadn't watched the video, but had read the STH articles about it.)

 

https://www.servethehome.com/wd-red-smr-vs-cmr-tested-avoid-red-smr/2/

 

This isn't some insane transfer. It's a NAS. It preforms backups and bulk storage. That is not ok for a drive that if you only look at WDs marketing and specs looks purely better than the CMR equivalent model.

Not saying the test is wrong but that is for a single disk, in a RAID set copying 125GB would result in a smaller performance delta but you'd still see it if you increased the sample size. Plus if you're using any decent backup tool you'll be doing incremental changes only and synthetic full backups (read last full + incs to create new full) or incremental forever backups (Veeam).

 

Pretty much all the low end SMB NAS's I've deployed have been for backups either for use with Windows Server Backup, Microsoft Data Protection Manager, VMWare Data Protection & Recovery or Veeam in the education market and these NAS's do nothing during the day and then during nightly backups only write a small amount of data because data change rates aren't actually that high even with 500-3000 students. Having a cheaper option here for disks that can actually be put in to RAID allowing longer data retention would actually be very nice, not every school can buy a 24 bay rackmount NAS or pay for equivalent cloud storage, there are ones that can but a lot more that cannot.

 

18 minutes ago, Curufinwe_wins said:

(Note I HATE watching videos when articles work just as well, I hadn't watched the video, but had read the STH articles about it.)

True that, also terrible for trying to source quote like here.

[leadeater]

7 minutes ago, Delicieuxz said:

Where there's the argument of 'If the change broke ZFS performance, then ZFS is the problem' going around, the designer comment of "WE DO NOT BREAK USERSPACE!" has relevance.

There is no such argument, that as stated was an example of flawed argument and logic to the kind being used. Nobody is actually saying this.

[Delicieuxz]

8 minutes ago, leadeater said:

There is no such argument, that as stated was an example of flawed argument and logic to the kind being used. Nobody is actually saying this.

No?

 

17 hours ago, leadeater said:

And WD's counter to that is ZFS is not all NAS's so not working with ZFS doesn't make it not fit for purpose.

 

16 hours ago, leadeater said:

*Sigh*, ONLY for ZFS.

 

16 hours ago, leadeater said:

I can prove ZFS not being fit for purpose using WD Red FAX because only it has a problem with that hardware. When you are the odd one out you are the problem.

 

16 hours ago, mr moose said:

Just to use a horrid analogy, if shell produced a fuel that worked in every car except Hyundia's,  you wouldn't say it was an issue with the fuel, you'd say it was an issue with Hyundai's.

 

 

That more than implies that the problem is ZFS and not the change to SMR.

[leadeater]

35 minutes ago, Shreyas1 said:

What I wonder is, will we need specialized judges for these types of cases in the future?

 

I mean, my impression is that the avg judge probably doesn't even know how exactly a hdd works or what NAS stands for, much less about different hdd technologies and such.

 

In the future, as these cases will become more popular, will we need specialized "tech judges" who know more about these things and will be able to make a fairer decision? Or will this negatively impact our justice system in any way?

 

just a thought

They are asking for a jury trial, plus judges are usually pretty decent at dealing with subject matters they are not experts in or have full knowledge of. It's something they would have to deal with a fair amount so it's really on how good of a case you can present that matters the most.

[Curufinwe_wins]

4 minutes ago, leadeater said:

Not saying the test is wrong but that is for a single disk, in a RAID set copying 125GB would result in a smaller performance delta but you'd still see it if you increased the sample size. Plus if you're using any decent backup tool you'll be doing incremental changes only and synthetic full backups (read last full + incs to create new full) or incremental forever backups (Veeam).

 

Pretty much all the low in SMB NAS's I've deployed have been for backups either for use with Windows Server Backup, Microsoft Data Protection Manager, VMWare Data Protection & Recovery or Veeam in the education market and these NAS's do nothing during the day and then during nightly backups only write a small amount of data because data change rates aren't actually that high even with 500-3000 students. Having a cheaper option here for disks that can actually be put in to RAID allowing longer data retention would actually be very nice, not every school can buy a 24 bay rackmount NAS or pay for equivalent cloud storage, there are ones that can be a lot more that cannot.

 

True that, also terrible for trying to source quote like here.

So I agree for semi-professional deployments, but I really was thinking primarily about end consumer situation with these drives and DYI or semi-DYI (expansion/replacement)

[leadeater]

12 minutes ago, Delicieuxz said:

That implies that the problem is ZFS and not the change to SMR.

That shows that two things together have a problem and should not be used, it doesn't prove that ZFS is not fit for purpose as a NAS and neither does that combination show that the disk is not fit for purpose as a NAS disk. It shows exactly what it shows, WD Red (FAX) + ZFS = bad, don't do it.

 

See the problem here is you wanting to make sweeping applications to everything rather than seeing the situation where something is a problem. One case is bad so all cases are bad, what kind of logic is this?

 

Edit:

Also nice out of context quote for the second one and fourth. How about you actually read the post otherwise you'll be ignored from this point.

 

Edit 2:

Actually no you have zero intention of taking what was written and the point so I will ignore, you know damn well what those posts were saying.

 

16 hours ago, leadeater said:

Yes this is a stupid argument but I can prove it using this method, it doesn't make it correct, literally the point here.

[Delicieuxz]

1 minute ago, leadeater said:

That shows that two things together have a problem and should not be used, it doesn't prove that ZFS is not fit for purpose as a NAS and neither does that combination show that the disk is not fit for purpose as a NAS disk. It shows exactly what it shows, WD Red (FAX) + ZFS = bad, don't do it.

That's just rationalizing the situation without explaining or justifying it. They have a problem after a change was implemented. The question is, does the entity which made the change to the drives without notifying anyone bear responsibility for those drives not working for some of their their excepted applications?

 

'I can prove ZFS isn't fit for purpose because it doesn't work following this change' is not proving your claim, it's just making the assertion.

 

1 minute ago, leadeater said:

See the problem here is you wanting to make sweeping applications to everything rather than seeing the situation where something is a problem. One case is bad so all cases are bad, what kind of logic is this?

Am I making sweeping applications, plural, or just one? And a sweeping application of 'don't break the userspace' sounds like a pretty good idea, to me.

 

But I'm not necessarily seeking to apply it. I'm just sharing a perspective that is missing from the discussion. It sounds to me like you're taking what I posted and going off into the wilds with it.

[leadeater]

1 minute ago, Delicieuxz said:

'I can prove ZFS isn't fit for purpose because it doesn't work following this change' is not proving your claim, it's just making the assertion.

Read the line below that, go on I dare you, or did you and just chose to ignore it so you could argue a point not made because you want to. Hence my reasoning for ignoring you, which will commence now.

[Delicieuxz]

Just now, leadeater said:

Read the line below that, go on I dare you, or did you and just chose to ignore it so you could argue a point not made because you want to. Hence my reasoning for ignoring you, which will commence now.

Hey, if you want to behave that immaturely, go ahead. I'm shocked at your behaviour, though. Bad day?

[leadeater]

Just now, Delicieuxz said:

Hey, if you want to behave that immaturely, go ahead. I'm shocked at your behaviour, though. Bad day?

Well sorry if people find out of context quoting and arguing back at someone with something they didn't say while acting like they said it annoys them. Seriously give someone the decency to read what they actually wrote and not just hand pick the parts you want and brush away what was said and the meaning. Why should I continue a discussion with you when you are doing this?

[Delicieuxz]

On 5/30/2020 at 7:50 PM, leadeater said:

Well sorry if people find out of context quoting and arguing back at someone with something they didn't say while acting like they said it annoys them. Seriously give someone the decency to read what they actually wrote and not just hand pick the parts you want and brush away what was said and the meaning. Why should I continue a discussion with you when you are doing this?

I don't know what you mean by you didn't say it. Even if you're playing devil's advocate or characterizing a type of argument, you're still presenting that side of things, like you did here:

 

On 5/30/2020 at 2:52 AM, leadeater said:

Like I said that doesn't prove that the WD Red FAX product is not fit for purpose as a NAS disk, it only proves it is not suitable for use with ZFS. You proved nothing on that. If I go out and buy 100 NAS's from different vendors with 8 or less bays, and I'll be generous here, 10 of them are ZFS based then only 10% would exhibit an issue and the commonality is ZFS. The other 90% that were not ZFS had no issue, so what exactly was actually proven here?

 

So, what's wrong if I present representation of the opposite approach, which also exists?

 

I wasn't attacking you when I posted the Linus Torvalds quote. I didn't quote anyone when I did that, I merely showed the argument - and doing that appeared to irk you. And when you claimed nobody had argued something relevant to that argument, I showed example arguments from this thread where positions had been presented that do relate to the Torvalds comments, and I quoted only the exact assertions which the Linus Torvalds viewpoint is counter to. I was not making an attack on you.

 

I think you misinterpreted the post I made featuring Linus Torvalds' comments.

[leadeater]

Just now, Delicieuxz said:

I don't know what you mean by you didn't say it. Even if you're playing devil's advocate or characterizing a type of argument, you're still presenting that side of things, like you did here:

 

40 minutes ago, Delicieuxz said:

Where there's the argument of 'If the change broke ZFS performance, then ZFS is the problem' going around, the designer comment of "WE DO NOT BREAK USERSPACE!" has relevance.

 

31 minutes ago, leadeater said:

There is no such argument, that as stated was an example of flawed argument and logic to the kind being used. Nobody is actually saying this.

 

You commencing to pull quotes from posts without reading the actual post or why they were posted.

 

Sample of the entire quote from one you used:

17 hours ago, leadeater said:

That's still a singular like I said. I can prove ZFS not being fit for purpose using WD Red FAX because only it has a problem with that hardware. When you are the odd one out you are the problem.

 

Yes this is a stupid argument but I can prove it using this method, it doesn't make it correct, literally the point here.

 

Edit:

I.e. You cannot prove WD Red FAX as not being fit for purpose using ZFS alone.

Second paragraph and the edit paragraph which also existed before you quoted it.

 

5 minutes ago, Delicieuxz said:

And when you claimed nobody had argued something relevant to that argument, I showed example arguments from this thread

Yes my quotes specifically used to highlight the usage of a bad argument, even stated so in the very post, and a post from someone else in reply to that who also understood what I said.

 

I don't need to claim nobody made the argument, I know nobody made that argument. I as the one who used it knows it is not as I posted it. If you had actually read the full post you would have known also but far as I can see you did not.

[Delicieuxz]

9 hours ago, leadeater said:

You commencing to pull quotes from posts without reading the actual post or why they were posted.

 

Sample of the entire quote from one you used:

Second paragraph and the edit paragraph which also existed before you quoted it.

 

Yes my quotes specifically used to highlight the usage of a bad argument, even stated so in the very post, and a post from someone else in reply to that who also understood what I said.

 

I don't need to claim nobody made the argument, I know nobody made that argument. I as the one who used it knows it is not as I posted it. If you had actually read the full post you would have known also but far as I can see you did not.

But you say you don't think the case has much merit to it because the issue only affects ZFS, and that you see that as a problem and an indicator that the fault therefore lies with ZFS and not the change in the drives:

 

On 5/30/2020 at 1:40 AM, leadeater said:

The problem I see is that ZFS is the only use case and software I have seen where the impact goes below the bar of acceptable product behavior and consumer expectation, within this there are two points of argument. First is that if everything else does not have an issue then does the fault actually lie with ZFS and is not a product fault with the hard drive.

 

I don't know how I'm interpreting that incorrectly.

 

Even if you want to claim that you weren't actually speaking your view and only that you see the possibility for an argument there (which isn't particularly clear in how you've worded things), you still represented that argument and so there is basis for the 'don't break the userspace' counter-argument to be presented.

 

 

Really, all I did was show that there's a counter-argument to the claim that if one thing no longer works while everything else does following a design change, then that thing that no longer works is the problem and not the design change.

[Shreyas1]

36 minutes ago, leadeater said:

They are asking for a jury trial, plus judges are usually pretty decent at dealing with subject matters they are not experts in or have full knowledge of. It's something they would have to deal with a fair amount so it's really on how good of a case you can present that matters the most.

With a jury trial, would the problem of uninformed people be magnified? I'm imagining a person who's already pissed about being picked for jury duty having to listen to this stuff, which IMO is quite complicated for the avg person.

 

Or is it a special jury? IDK how this works tbh

[leadeater]

10 minutes ago, Delicieuxz said:

But you say you don't think the case has much merit to it because the issue only affects ZFS, and that you see that as a problem and an indicator that the fault therefore lies with ZFS and not the change in the drives:

That's odd because I clearly remember saying this has a 60% chance of succeeding.

 

18 hours ago, leadeater said:

Given that cases like these tend to favor the consumer I think I'd give it a 60/40 shot at winning.

This also happens to be the very last line a post you used a quote from. And I highly suggest you actually read the entire post from start to end because what I was talking about from the bits you are quoting is what I believe WD will attempt to argue in court.

 

Thing is my opinion on the matter, one of many, is that factual evidence matters and so does good argument structure. I have a problem with people stating opinions as if they are facts without the evidence to support the claim. If people want this case to succeed and for WD to be held liable then you need better evidence than people screaming on Reddit and ignoring of test data that actually exists that shows something contrary to their belief. 

[mr moose]

1 hour ago, Curufinwe_wins said:

Ok so let me be blunt. We hashed this out in great detail in the original thread about the issue being brought to awareness, and I honestly don't have the time nor inclination to dredge back up all of the sources and evidence discussed at that time. If that means you disbelieve the previous discussion, I understand, even if it's annoying.

 

I do have this bookmarked ofc, and it's from around that time. It shows at least a few random incorrect employee statements and other situations. 

 

https://arstechnica.com/gadgets/2020/04/caveat-emptor-smr-disks-are-being-submarined-into-unexpected-channels/

In the original thread I only maintained that we do not have enough data to blame WD completely or that there was sufficient data to claim all other SMR disks did not fulfill their intended use scenario.  Since the majority of that discussion we have now had proof that WD disks do in fact work as intended and the penalty for being DM SMR is not as bad as everyone was making out.  Now, certain moderation occured and that thread was used as one of the reasons for it,  I don't know exactly what I did that was so terrible other than support my views (which have been more than vindicated now) against aggressive accusations of shilling, so I am reluctant to go into too much more detail other than to to point people to the video's leadeater has provided.

 

 

[Delicieuxz]

On 5/30/2020 at 8:16 PM, leadeater said:

That's odd because I clearly remember saying this has a 60% chance of succeeding.

That's a middling chance - which is to say you don't see it as a strong case.

 

You also called the people bringing forth the case:

On 5/30/2020 at 4:50 AM, leadeater said:

Literally just some salty people

And identified with the position that ZFS could be the problem by sneering at the plaintiffs:

On 5/30/2020 at 4:50 AM, leadeater said:

likely using the product out of spec anyway

... and implying they're arrogantly behaving as if:

On 5/30/2020 at 4:50 AM, leadeater said:

their usage represents an entire market.

 

Those aren't impartial statements.

 

But in all of this, I think it seems as though you took my presentation of Torvalds 'don't break the userspace' commentary as a personal jab at you and assumed things about what I was saying, when what I meant by it is simply a counter to the presented argument of 'the odd-one-out is the problem, rather than the change which created the odd-one-out'.

 

I think this has been blown out of proportion. My post featuring Linus Torvalds' commentary is adding additional perspective to a line of thought presented in this thread's discussion, and is not saying something about you or other people for presenting or exploring that line of thought.

[leadeater]

1 minute ago, Delicieuxz said:

Those aren't impartial statements.

Of course those are not impartial statements, and I very much believe they are just being salty. Because as stated in that post or the one after it this case is seeking in restitution a refund of the product or replacement, something WD is already offering, and demanding the WD stop marketing them as NAS ready (because their usage with ZFS had a critical problem and definitely should not bed used together). This is why I'm saying the are just being salty because point 1 is already on offer and point 2 they are acting like their usage represents the entire market (bad argument and too easily countered).

 

As with the other thread on this I agree DM-SMR should have been on the product sheet, I have some idea why WD wouldn't want to put it there though.

 

6 minutes ago, Delicieuxz said:

But in all of this, it seems as though you took my presentation of Torvald's 'don't break the userspace' commentary as a personal jab at you and assumed things about what I was saying

Not really, at least not by itself, it was the usage of my quotes in a manor to backup a point you made when they weren't at all what you were using them as.

[Delicieuxz]

11 minutes ago, leadeater said:

Not really, at least not by itself, it was the usage of my quotes in a manor to backup a point you made when they weren't at all what you were using them as.

I am not sure what you saw me as using them for, but I was using them just to show what comments (whether they're yours or devil's-advocate comments is pretty irrelevant to the Torvalds commentary, from my point of view) in this thread the Linus Torvalds commentary is relevant to, following your suggestion that the commentary wasn't relevant.

 

Maybe I could have made the intended purpose of my initial comment clearer. I thought it was humorous, though, just showing Torvalds spazzing out over the matter.

[wanderingfool2]

10 hours ago, leadeater said:

Yes, all the videos are in the original topic about this, it's about 5 video series going through new array setup and rebuild to full data array rebuild with mixed disks and all FAX.

So, I found the videos and watched them and the data doesn't really show anything in defense for SMR's being "NAS" grade.  He got his maths wrong, and he never truly did a test of CMR vs SMR drives (just varying degrees of amount of SMR drives).

 

10hours 40 min (640 minutes) [CMR drives with single SMR]

18 hours 2 min (1082 minutes) [All SMR drives]

 

The statement in the videos is 59.15% slower, that is incorrect [Correct math is (1082-640)/640 = 69.1%].  8 hours is a considerable difference in rebuild times, and trust me it would make a difference in my purchases if I knew a rebuilt would take 8 hours longer on an normally 11 hour rebuild.

 

The next issue is the testing methodology.  With STH, they used the same NAS, swapped the drives, and emulated real world conditions of data being written while resilvering (I like this methodology).  In the NASCompares video, it was different NAS, they only ever did swap the AX drives, and they put 0 load on it.  In the real world, if you had to rebuild the 8 hours of differences makes a world of difference (in any workplace environment)...if they used the drives while rebuilding (a likely scenario since a business can't shut down to have it rebuild); then that 8 hours would likely balloon a lot more with SMR drives (since the cache would be used up, and since it no longer is writing sequentially it will likely suffer a lot more)

 

10 hours ago, leadeater said:

All times were less than 24 hours, 40% looks big if you remove the context.

Less than 24 hours in an ideal rebuild scenario, see above

 

10 hours ago, leadeater said:

Rebuild times are not the be all and end all of everything, that is also silly. If it is high enough where it actually is a problem, less than 1 day to 9 days, then it is a problem.

 

Edit:

FYI on engineering a result to prove a bad point, it takes about 3 days to rebuild a 3TB NL-SAS 7200 RPM disk in our Netapp that is used for backups because of the consistent load placed on the disks. 3 days is a very long time for a 3TB disk, it doesn't make it not fit for purpose so I'd never use this to try and argue that sort of line.

 

The additional context that needs to be applied here is that each disk shelf is configured in triple parity with two hot spares and there are two Netapps in different cities and data is replicated to both, risk of data loss due to disk failures is very low and these longer rebuild times are not a significant issue.

It isn't about engineering a result, it all comes down to the fact that WD didn't including any indication that they silently switched to a technology that would have an impact on a metric that is important.  If you see my other posts before, I have talked about how SMR has it's perks, but in a NAS advertised drive it should be in the specs; ontop of which they keep the product name the same.

 

And yes, 8 hours differences can make a world of difference in real world rebuild cases.

[mr moose]

44 minutes ago, wanderingfool2 said:

 

 

10hours 40 min (640 minutes) [CMR drives with single SMR]

18 hours 2 min (1082 minutes) [All SMR drives]

 

The statement in the videos is 59.15% slower, that is incorrect [Correct math is (1082-640)/640 = 69.1%]. 

https://percentagecalculator.net/

 

59.14972%

 

Some people will not be happy with that, absolutely.  But putting this into perspective, rebuilds like that are not weekly events, they are not even monthly events,  for some people they are yearly events. So the trade of is an extra 8 hours of rebuild time for how many usable TB the rest of the time?  

 

I get that people are pissed, but is it a mountain or a mole hill?

[leadeater]

5 hours ago, wanderingfool2 said:

.if they used the drives while rebuilding (a likely scenario since a business can't shut down to have it rebuild); then that 8 hours would likely balloon a lot more with SMR drives (since the cache would be used up, and since it no longer is writing sequentially it will likely suffer a lot more)

I already know this and stated so in the original thread about this.

 

5 hours ago, wanderingfool2 said:

In the NASCompares video, it was different NAS, they only ever did swap the AX drives

Then I don't think you watched many of the videos or paid too much attention to them. Two different NAS's were used with exact same test conditions and disk configuration to see if different NAS's acted differently, if there was a discrepancy outside of the hardware resource difference of the NAS's. They tested all CMR, Mixed CMR + DM-SMR and all DM-SMR for every test carried out.

 

Each sample required to be tested requires a very long time, at some point you just have to cut it where you can see the trend or you have enough data to form a conclusion. This could be correct or not, or miss what may actually be an important consideration not thought of at the time. You still have to remember all these tests took days to carry out.

 

The only testing missed out was generated client load under a rebuild. The only other one I can think of is pretty much irrelevant or covered in the other tests, rebuilding a CMR disk with a DM-SMR disk in the array but that wouldn't have any impact at all unless you did the generated client load along with it, and the rebuild I/O for the DM-SMR disk would be read only anyway.

 

I seriously question, from actual experience, how much load people actually put on these 8 bay and lower NAS systems. I don't disagree at all with the DYI FreeNAS situation at all, never have but I'm not going to try and argue that due to this DM-SMR disks are therefore not to be used in any NAS under any situation ever for all of time.

 

5 hours ago, wanderingfool2 said:

It isn't about engineering a result, it all comes down to the fact that WD didn't including any indication that they silently switched to a technology that would have an impact on a metric that is important.

Exactly which in my longer post (on the first page) about how I think it would play out in the legal setting is exactly what I have said. Also as I covered I also suspect WD did zero testing of the DM-SMR revision under ZFS and believed the difference in performance characteristics was not note worthy enough for them to place the change in recording technology on the product sheet. Hindsight easily shows that was clearly wrong and from a defensibility standpoint putting it there covers your ass just in case, WD didn't do that.

 

What I believe is the argument that actually does have a likelihood to win in court, I think it's the same for you, is that not disclosing this information on the product sheet resulted in consumers buying a product they would otherwise not have, misleading advertising. If they stick to that it is in my opinion likely to succeed, unlike the other points they have in the filling documentation.

 

Going in and getting a court to enforce that WD, there by any vendor, cannot advertise a DM-SMR disk as suitable for NAS usage is just a bad idea, I have no idea why anyone wants this.

[wanderingfool2]

6 hours ago, mr moose said:

https://percentagecalculator.net/

 

59.14972%

 

Some people will not be happy with that, absolutely.  But putting this into perspective, rebuilds like that are not weekly events, they are not even monthly events,  for some people they are yearly events. So the trade of is an extra 8 hours of rebuild time for how many usable TB the rest of the time?  

 

I get that people are pissed, but is it a mountain or a mole hill?

The last calculator on the website is the correct one to use (which is the 69.1%).  Otherwise I could claim a horrible drive is only 0.001% slower (e.g. if I have a drive that rebuilds in 1 day, and another that rebuilds in 100,000 days I could say it is only 0.001% slower when in actuality it's 1,000,000% slower)

 

The way I look at it, it is to ensure companies don't get complacent with these kind of things.  For some, yes the trade off is okay, but it isn't up to WD to decide for the consumers (and pretend there isn't any difference).  No one seemed to notice as well I was writing 8 hours either, when in actuality it was closer to 7, which is why it is always important to look at the raw values.

 

3 hours ago, leadeater said:

Then I don't think you watched many of the videos or paid too much attention to them. Two different NAS's were used with exact same test conditions and disk configuration to see if different NAS's acted differently, if there was a discrepancy outside of the hardware resource difference of the NAS's. They tested all CMR, Mixed CMR + DM-SMR and all DM-SMR for every test carried out.

I had watched the rebuild of the 90-95% rebuild, and the write test. In the rebuild test at full they actually compared mixed (single SMR) and pure SMR.  I didn't watch the remaining videos, as I really don't like his testing methodology (on top of using the wrong calculation for the metrics, and yes I know I was say 8 hours when it's closer to 7).

 

3 hours ago, leadeater said:

Each sample required to be tested requires a very long time, at some point you just have to cut it where you can see the trend or you have enough data to form a conclusion. This could be correct or not, or miss what may actually be an important consideration not thought of at the time. You still have to remember all these tests took days to carry out.

Yes, it takes a long time, but the point is I think the methodology produced the wrong data (due to not properly testing the correct initial conditions).  If I were doing the testing of a 90%, I would roughly done it as follows:

CMR full (on NAS with the same specs, which they had at 90% full): rebuild both, but on one, run simulated workload

Repeat for SMR full

The SMR, my guess would take significantly longer.

 

While the difference might be only 7 hours, what if WD took that approach with their 8TB drives or 10TB drives (it becomes 14 hours).  I am not saying they shouldn't be allowed to, just that they need proper disclosure of something that could have such a large impact on customers

 

3 hours ago, leadeater said:

I seriously question, from actual experience, how much load people actually put on these 8 bay and lower NAS systems. I don't disagree at all with the DYI FreeNAS situation at all, never have but I'm not going to try and argue that due to this DM-SMR disks are therefore not to be used in any NAS under any situation ever for all of time.

I've seen 4 bay ones filled to nearly 100% before in the wild.  Actually, now I think of it...I've actually seen a rebuild before matching that (albeit 2TB drives at the time).  I'll agree, SMR's can fit into a NAS situation (they can be cost effective, especially if you don't have to worry about rebuilds in the future).

 

3 hours ago, leadeater said:

What I believe is the argument that actually does have a likelihood to win in court, I think it's the same for you, is that not disclosing this information on the product sheet resulted in consumers buying a product they would otherwise not have, misleading advertising. If they stick to that it is in my opinion likely to succeed, unlike the other points they have in the filling documentation.

 

Going in and getting a court to enforce that WD, there by any vendor, cannot advertise a DM-SMR disk as suitable for NAS usage is just a bad idea, I have no idea why anyone wants this.

We agree on this point quite strongly I think.  [I think it's more people upset that it was hidden by WD in NAS drives and that by doing so it feels like WD seems to think SMR doesn't have any impact on NAS systems *a guess at the reaction*]