Less endurance than cheap SSDs??? WD Red Pro 20TB HDD

[@christopherjackson2157]

If it's not suitable for use in nas appliances, just don't market it as if it is. Remember the Seagate archive drives? I had no problem with them being crappy because it was clearly acknowledged in the marketing. My problem with wd isnt their products; it's their deceptive marketing practices.

[@robcrawford6083]

This guy is being very deceptive. It’s 300tb written per year and it’s standard. I have looked at other nas drive manufacturers and they say the same thing, 300tb a year. If you have a single 20tb nas drive you would have to completely rewrite the whole thing 15x every year, 5yr warranty = 65x300. If you have five 20tb nas drives striped you would have to rewrite 1.5 million gigabytes every year. Fuck this deceitful bastard.

[@1armbiker]

I'd just like to point out that if you have these in a ZFS system and do one scrub every week throughout a year, assuming it checks the whole drive that puts you at over 3x what the rating is just for data integrity checks, and if it only touches data you've put there, at 1/3rd capacity the scrubs would still eat the entirety of that workload allotment which is just disgusting.

[@arthurmoore9488]

Put another way, if you only did a ZFS scrub every month with a full drive, then you could only read the drives contents for actual workloads 3 times! 300 - (12*20) = 60. That's assuming the data was already on the drive to start with.

[@kelownatechkid]

@@arthurmoore9488 I scrub every pg in my ceph cluster every 90 days or less, this alone makes these drives a terrible investment!

[@savvassidiropoulos5952]

I just changed the scrubbing schedule on my nas from once a month to once every 3 months.

[@WorBlux]

@@arthurmoore9488 You forget write amplicatation. A 5 drive raidz1 has a 25% read/write amplification. (More in reality because ZFS doesn't force a full width allocation. though the new draid format does. In which case one scrub a month hits the workload rating with zero leftover. However I think this is just a conservative performance/design guide. The datasheet notes that " Workload Rate will vary depending on your hardware and software components and configurations." And a zfs scrub is basicly the ideal workload for a hard drive. So long as your data pool has enough of an inactive data period in it for the scheduled scrubs you should be fine. (keep in mind a scrub on these drives takes over a full day, so some script-fu/auto-scrub with pause + resume may be needed anyways to avoid filling the drive cache with scrub activity rather than the zfs filesystem metadata that you actually want in a cache to accelerate actual pool/filesystem use by applications or users. (Although ZFS on a small NAS probably has more data in it's ARC cache than the drives will keep and it's smart enough not to overwrite this with a scrub, so I'm probably overthinking it even here) THE TLDR: Until someone actually reports problems I wouldn't worry too much, ZFS is awesome, but keep in mind large HDD's have severe bandwidth and I/O limitations compared to capacity. If you have a 10 bay NAS and the workload is more than 3 PB/year consider more spindles, better drives, or a tiered storage approach. (SSD caches)

[@andyeban]

what is a scrub and should I be doing them on my 16tb NAS?

[@drchanas]

Instead of being scared each time about where the "gotcha" is hidden in the WD spec sheet and when will it bite me, I just go Exos.

[@kungfujesus06]

Seagate has spooked me enough with their last few lines of 7200.x/barracuda drives that I can't trust them

[@beaker-yt]

I also use Exos since 2+ years and have no problems with them. But they too have a rather bad UBER of 1 per 10^15 and an annual workload rate of 550TB/year. If you exceed that workload, they do not guarantee the MTBF (2500000h) and AFR (0,35%).

[@kungfujesus06]

@@beaker-yt it's not the same product line, I know, but there seems to be several severe firmware bugs on this line every product release en.wikipedia.org/wiki/Seagate_Barracuda?wprov=sfla1 That's not even mentioning mechanical failures, which seem to be higher with Seagate.

[@gentuxable]

imho no manufacturer is that safe that you can afford to not have parity (RAID) in the first place and nowadays it would be extremely rare that mulitple drives fail within a short timespan as you wouldn't be able to recover.

[@kungfujesus06]

@@gentuxable in my experience that doomsday scenario has been more likely with Seagate.

[@chronowerx]

So they're drives with no warranty really... If you buy a 20TB, you have a need for that much space, so let's say you fill it when you get it and your system does a full read for verification every month - your drive is out of spec for warranty in less than 6 months, even if all the user has ever done is fill it once and never actually accessed a file from it. That's crazy. Nearly every NAS has a patrol read / scrub / integrity verification, so WD are markeing something not fit for purpose. Stick a blue or green label on them if they're this bad!

[@the_beefy1986]

This reminds me of the cheapification of the Black drives in the 2009-ish timeframe; WD decided to disable TLER on the black drives to force RAID users to go to the "Enterprise" models, which of course cost 2x-3x as much per-TB. It was well-known in those days that the only real difference between these product lines were tweaked firmwares. I immediately took a disliking to WD and started using HGST and Samsung drives.

[@hescominsoon]

Hgst is now wd

[@the_beefy1986]

@@hescominsoon I understand that. That didn't happen until several years later.

[@gravesclay]

"I immediately took a disliking to WD and started using HGST and Samsung drives." - followed immediately by above average failure rates.... NEVER samsung. Not even once. Worst disks I've ever used other than Seagate.

[@the_beefy1986]

I think I had 2 Samsung drives in my RAID6 array shortly after. I don't recall them being particularly worse than the others I had during that era. Remember we're talking about more than a decade ago. Things can fluctuate quite a bit over that time.

[@whitebeartigtig]

I still have a couple old HGST drives, one of them has about 8 years of power on hours. too bad drives this good aren't made anymore.

[@MoraFermi]

300TB/year is ~9.5MB/s. I wonder if they have been copy-pasting the stats in those brochures since ATA/66 days. On a side note, I wish I could get a cheap "consumer grade" 20TB SATA SSD...

[@beauslim]

It seems like short warranties on expensive items is becoming a thing. Companies have noticed that people aren't looking at that when purchasing. It is really short sighted because they burn their brand to nothing with people. My dad bought a Samsung fridge and it died after 14 months. 1 year warranty, and no parts to be had. He'll never buy a Samsung anything ever again.

[@CPUjunkie]

I've had too many hard drives die on me. I noticed the ones that died the quickest were the ones with the smallest warranties. I just got a 12tb drive so did a lot of comparing. So many drives/websites just have a 1 year warranty. Some were 2 years. You gotta look to find the 3/5 year warranty

[@depth386]

Oof but Samsung SSD’s are great

[@SpyderBlackOfficial]

This reminds me of Samsung phones. Every Samsung phone i had died 2 or two months after warranty. Three phones to be exact. Third time was my final straw. Boy was I nieve.

[@thedopplereffect00]

I always try to buy hard drives with a 5 year warranty

[@theosexpertdaymon2774]

I've never seen a damn thing one from Samsung I liked. Bloated half-assed janky software on TVs and phones. And the stuff seems to live 2 years at best.

[@sunxore]

I have kind of lost trust in all disks from all manufacturers so now I just buy a drive with CMR and long warranty and hope for the best (and use data integrity checks and backups). Last disk bought was a Seagate Exos enterprise drive and it has been really good so far (and cheap). In practice my drives usually survive a long time.

[@annihilator9075]

I only use the EXOS drives... I am betting they'd be a pain to send in for a warranty if it's possible at all, but so far they have been solid. With these kinds of shenanigans, companies are pulling with low-performance drives, they probably wouldn't accept a warranty from a well used one anyways.

[@Elazarko]

I want to get my first NAS can you please recommend me what to get? I'd like about ~10tb and not looking to spend too much as long as it's reliable..

[@Banner1986]

I just gave up on WD completely when the whole SMR/CMR debacle happened - I no longer care to deal with them as a company regardless of how good a product they may or may not have as long as a viable alternative exists.

[@charleshines1553]

They should never ever have sold SMR drives as a NAS drive which they did do and were sued for rightfully so

[@titan_fx]

Yeah, I just full stop on WD. I don't support their business model since SMR Red NAS debacle.

[@henfibr]

Conclusions: This translates to about 0.05 DW/RPD (5% Drive combined writes/reads per day) which is one of the lowest ever and certainly not suitable for pro NAS usage with data scrubbing, background tasks such as thumbnail generation, and other read/write amplification functions. Also, running badblocks (r/w mode) on this 20TB harddrive before first usage will spend 15% of the Workload Rate Limit. On other news, I just checked and Ironwolf Pros have the exact same WRL for the 20TB and the whole lineup. I also found a 2016 PDF from Seagate, that explains the reasoning for the introduction of this Workload Rate Limit in HDDs and how drives after this threshold become less reliable in a non-linear fashion (up to this limit, the reliability is about the same regardless of the amount of data read/written). Using multiple smaller disks would be a way to increase the aggregate WRL for read/write intensive operations (although TCO may be higher eventually)

[@annihilator9075]

looks like they don't have the WRL for their EXOS class. I couldn't find it... I also didn't see it for the ironwolf until I looked in a different section...note: it's not under reliability... I found it in the first category they had on the data sheet. Is this metric only used to reject warranty claims or does it also apply to enterprise drives just not mentioned?

[@henfibr]

@@annihilator9075 You're right that WRL is missing from the EXOS datasheet, but it is listed in their landing page when you click "Features", which says (verbatim): "Enterprise-class drives with a 550 TB/year workload rating and 2.5M-hr MTBF to meet the most demanding storage requirements." Since it is missing from the datastheets, It seems like a "recommendation" for the EXOS rather than a specification/requirement for warranty claims. That said, at least the Seagate drives (both Ironwolf Pros and Exos) list 1 sector per 10E15 as non recoverable read errors per bits (compared to 1 per 10E14 for the WD Red Pro).

[@robertslavin2557]

@@annihilator9075 I believe the EXOS are

[@TheExileFox]

This only helps if there's actually a cheap and reliable option for choosing a NAS unit. It basically does not exist. I'm running an ancient 2bay Zyxel and it's alright. I'm looking at maybe getting their ancient 4bay box off ebay because gnap/synology second hand is about as expensive as buying the brand new unit.

[@Banner1986]

Oh I know! They saw ISPs getting away with "1Gb/s download - 1TB data cap" and went, "... why didnt we think of that!?"

[@RN1441]

While I'm no fan of WD, and avoid their products after the SMR concealment, Black series downgrades, and other shady business decisions, I think an accelerated life cycle test is in order. Set up a consumer SSD of your choice and one of these red drives to move data back and forth at full speed and record the metric for time and data moved until they encounter an unrecoverable error. I'm suspicious that the SSD is actually going to die first regardless of what the rating says.

[@faithful451]

This

[@mcouture8169]

Noob and hobby person only here, but didn't he say that the realistic situation is more "reads" than "writes". Doing a full write/read each time favors the HDD, while the SSD may work best in a single "write" followed by 99 "reads" - perhaps like a YT video or such. I apologize if this is pure idiocy... just seeking to understand as best as I'm able to do so. Have a great day!

[@wayland7150]

It's only the huge capacities that keep hard drives viable. If they had stopped getting bigger at 2TB then everything would be SSD by now.

[@TheBackyardChemist]

Here is a suggestion: buy 5 of these drives, retail, put them in a test bench with an IO benchmark in an infinite loop. Lets see if these drives really do fail after 300 TB, or if WD is full of it.

[@quag]

WD are using an “up to” number. So they’re saying that all 5 drives should fail before hitting 300 TB in a single year.

[@jamescollins6085]

@@quag For a drive of that capacity, 300TB seems like next to nothing. That's only 15 full writes, if my calculations are correct.

[@ATGEnki]

@@quag Not quite. If my understanding is correct, they warranty the drives for 5 years, so it would be 1500 tb is about when the drives should start having issues. They just use really low numbers to justify a shitty product.

[@adamreid5901]

Too small a sample size mate

[@gleep23]

That is not the point. If you exceed their advertised specs you cannot expect to make a successful warantee claim. When making a long term investment on several devices, you need a warantee to protect you from unexpected & costly replacement.

[@tcntad87]

How and why you only have 81K subscribers I have no idea, you deserve alot more!

[@KillaBitz]

I feel like that should be in the desktop line and the Red Pro should be a 550tbw like the rest. The Red Pro is not marketed as a budget product so people would be upset to think they were getting the data centre rejects. (time to check the data sheet on my 16tb units)

[@devinbaines4952]

I got hit with the whole SMR debacle as I was putting together a NAS in early 2020. After a couple of RMA screwups, I wound up with Red Pros as replacements. I thought I lucked out. Perhaps not, it turns out. Anyway, in addition to getting Red Pros to replace the SMR drives I had inadvertently bought, I am also getting $9/drive in settlement money from a class action suit brought here in Canada on behalf of purchasers of SMR drives, so there's that.

[@ServeTheHomeVideo]

Whoa! Big $$$

[@ikjadoon]

Glad to see some settlement money came out. I love reading the silly settlement conclusions: "Western Digital denies these allegations, denies that SMR is inferior technology, or that it did anything wrong. "

[@devinbaines]

@@ikjadoon I don't have the money yet. Only very recently was I asked to provide the purchase invoice and drive serial numbers to the law firm handling the case. Fortunately, I still have all the emails from the RMA process containing that information. These things take time.

[@TheDillio187]

9 dollars to the consumer, and probably 90 million to the lawyers

[@kattz753]

I am SO disappointed with WD. I bought a couple of 4 tb Iron Wolf drives yesterday. The 6tb WD Red + was only $20 more. That adds up even when buying 2. I just don't trust them anymore. And I'm still running 2 15+;year old Greens. They're going now even though there are still NO errors on either one. It's going to be a mechanical failure any day at this point. These things do go in cycles. Back when I bought those drives you couldn't trust Seagates. They were failing like crazy. We called it the death crunch. Everyone I knew were having babies. I had panicked people at my door with hard drives in hand begging me try to get all their baby pictures off the drive. Tiger Direct was literally giving away Seagates with system builder copies of Vista just to get rid of them. That was ironic.

[@Alan_Skywalker]

A little correction: Usually only enterprise SSDs rate their endurances with JESD219 or pure 4K random standard. Consumer SSDs' endurance ratings are usually sequential writes, with controllers recording the NAND writes directly. It's not really practical to rate consumer SSDs with enterprise standards. They have much less OP space, making the WA much larger. When combined with simpler controller and firmware, lower grade and less tuned NAND flash, and mechanics like CTL/flash overclock and SLC cache, they won't last very long under pure 4K write workload.

[@someoneoncesaid6978]

If you're pushing through a petabyte of data in two weeks, you'd wear out the SSDs a lot quicker than you would a physical hard drive. I do video editing, so do a ton of reads/writes to the drive in the process. I tried using SSDs for this in order to get faster reads/writes, and burned through two SSDs before going back to using physical discs even though they're slower, because I've never worn out a physical drive while doing the same work.

[@acelee5732]

Nice name

[@NeptuneSega]

But SSD’s are physical as well. You prob mean mechanical hard drive instead

[@someoneoncesaid6978]

@@NeptuneSega - Yes, I was referring to the platters within mechanical drives when saying "physical discs".

[@jimmybrad156]

@@someoneoncesaid6978 Yeh, SSD's don't have any discs :)

[@redtails]

As for the reason why a 20tb drive has the same endurance as a 2tb drive... well it's the same motor, the same arm, the same write head, probably even the same platters. I always assumed that, aside from some models having more platters,/heads the "capacity" is a firmware setting. If the write heads have a max rated throughput, well there you have it.

[@leexgx]

Doing a monthly smart extended scan and a data Scrub every month is approx 444tb reads per year on a 20tb disk (18.18tb actual),, more if it's a Synology using btrfs with checksum enabled on all share folders, as it first scrubs the btrfs filesystem to verify data and correct if required, once finished a raid rsync is ran

[@philliumo]

It might have been helpful to compare, not just to other WD drives, but to what other manufacturers are offering. Sure, it might kinda suck as a product, but is anyone else doing significantly better?

[@majstealth]

if i not totally fucked up seagate says the ironwolf pro 20TB is good for 8.760h/a 285MB/s, which should result in 8571TB/a - that is quite a difference to be clear, i do not want to endorse seagate, i used them for decades and they fucked me over with smr also, i just wanted to state what i read of the datasheet

[@RJ_Cormac]

Seagate has the exact same ratings for IronWolf and EXOS

[@majstealth]

@@RJ_Cormac you are right, under the not so logical "Multi-User Technology (TB/yr) 300"

[@TheFourthWinchester]

@@majstealth Seagate has never fucked customers with SMR drives marketed as CMR like WD.

[@engineeringVirtue]

​@@TheFourthWinchester they've done worse...... something like 1/4 of the Seagate enterprise drives were failing very fast around the 2014-2018 timeframe and Seagate was taking forever to send replacements. Seagate basically gave the small business NAS market to WD. Before that, they had a great reputation developed over 25yrs pioneering the storage space. Keep in mind it was Seagate best drives that were failing, not their cheap ones. Toshiba and Fuji also grabbed much of the enterprise drive space from seagate. WD may be cheapened out on their lower end drives, but the WD Gold drives mostly have a flawless reputation. The only reason everyone isn't using them is noise and price.

[@Chloiber]

If you think about it, even in a Synology NAS, you typically do a disk scrub every month (also on other type of RAID systems or controllers). This goes through all data on a disk. If you use 20TB disks, you would go through 240TB of data a year...

[@annihilator9075]

I have around 14TB on my RAID right now with 16TB drives... I also do weekly scrubs, so I am looking at these numbers and there is no way I would be able to ever waste $$ into these drives unless they are ridiculously cheap then I could find a use for them. If they aren't

[@gleep23]

This is so sad. I was a loyal, 100% Western Digital supporter for decades. I was angry when the Reds quietly became SMR, but after they labelled things honestly I forgave them. I still didn't like that *any* Reds remain SMR. But now Red Pro with low endurance? That's it! I'm done with WD! I cannot trust the brand name any longer. I'm really upset, because I never had a bad experience, and I'm mid-warantee with many drives still. For future purchases, I'll be inspecting carefully across the market, all brands, checking, specs, warantee and reviews. Thanks for the heads up. Goodbye Western Digital. 😢

[@davidespada01]

what do you recommend ultrastar , Toshiba ?

[@JGoodwin]

I think they're driving down the warranty to improve margins.

[@Arachnoid_of_the_underverse]

Sounds to me like cost reduction on both quality and warranty.

[@firenado4295]

bruh i miss when drives lasted for way longer, my fathers voip server has a 20gb segate drive from 2001 and its been in that voip server for the last 17 or so years and it still works fine no errors. also have a wd 160gb drive from 2006 thats got way less useage but still running perfect i bought a bunch of used wd 80gb drives from 2007 for stupidly cheep (knowing they would be crap and i just wanted them for testing) but some of those are still going strong too. then there is my brand new segate 1tb hdd guess how long that lasted? 6 months its pitiful the way things are going. why should i have to buy a datacenter grade drive just for personal use if i want something thats not gonna die in 5 minuits

[@gravesclay]

"brand new segate" well.... there's your problem.

[@VraccasVII]

Keep in mind that while this happens sometimes, most of the brothers and sisters of that drive have long died. It can happen. But it's a gamble now as it was a gamble then.

[@firenado4295]

@@VraccasVII true i know 20 years for a drive is a miracle but with smr and cheep quality and low durability right off the bat drive longevity is just crap, manufacturers expect you to replace the drive every 5 years or so just because the paper says your warranty will end. my segate only had 2 years of warranty and it didn't even last that, had the exact same drive but the oem version in my mothers pc and that lasted 3 years before dying and thats still bad for a drive her old pc is 14 years old and still works fine, i do hope hdds dont go the same way printers went where it just comon place to through it out every couple of years just because the manufacturers make you

[@firenado4295]

@@gravesclay i mean alarmbells should have sounded in my head when i saw it was the cheepest deal on amazon but eh, also remembered my mothers new pc had the same drive and that only lasted 3 years

[@Hogdriva]

I have a 20mb Seagate drive from 1983 that is still kicking lmao

[@HHX_H]

I use gold even for my home pc it's not that more expensive ! I work in a Datacenter and the most drives that are scraped daily are ironwolfs and not WD ....

[@ServeTheHomeVideo]

Also amazing 1M hour MTBF for the Red Pro v. 2.5M hr for the Gold. I would say you made a good choice.

[@nighthawkvc25a]

Sometimes the Gold class is the same or slightly lower priced too. A Little over two weeks ago, Western Digital's store website had a sale for 2x Gold 16 TB drives for $600 total and it included a free 64 GB wireless sync charger. I was hesitant to get it because I wasn't sure of the differences between Gold and WD Red Plus/Pro, but am glad I did 😄

[@craigshea2930]

I have never seen a Red Pro drive priced lower than the same size Gold drive (excluding any sort of special sale). And, even if a Gold drive did cost a few dollars more than the Red Pro, you get a much better warranty, much higher MTBF and nearly 2x the WRL. I just don't see the value in the Red line--at all.

[@engineeringVirtue]

Gold drives make loud clicking noises every 5 seconds or so that can be annoying in a NAS if not in a high noise location. They also tend to run hotter requiring more cooling and higher fan speeds. So, if you want performance/reliability ... get the WD gold. If you want low noise/reduced cost of operation for smaller workloads... go red plus. I don't see the point of the red pro line at all...

[@Luscious3174]

300TB/year is roughly 820GB/day. The ONLY time I remotely approach that number is when I do my full system backup on the 1st of each month. I have a 10TB red drive exclusively for backups. I do incremental daily backups of my OS, software and data drives that rotate with a full backup of each every 14 days. Most days I barely have activity and the drive with the most stuff on it is under 2TB. Of course, compression plays a big part in backup software so the wear is minimized. If you want wear endurance get a 1.5TB Optane drive - those things will do 30x drive writes per day (45TB every 24 hours) making them phenomenal for a scratch disk, although they command an eye watering price. Alternatively an add in PCIe card with four m.2 sticks could also be doable. The largest U.2 drive at 30TB capacity will even give you a full drive write per day. And if that's STILL not enough there's always the option to go RAMDISK. It's all about picking the right tools for the job.

[@ServeTheHomeVideo]

The challenge is when you scrub disks. It is not just user/ client initiated tasks

[@cmuller1441]

The WD Red didn't include that limit previously! Just search "wd red spec sheet 2015" Is there some kind of flash cache like on hybrid HDDs ? This could be the #1 explanation of that limit... Maybe there are actually using SMR and hide it like that !?

[@cmuller1441]

Just searching for Optinand and... I was right... It's probably just the wear of the cache !!!

[@tonicipriani]

This had me go and check Seagate too, but apparently their Ironwolf Pro also has the same 300TB/year workload rate.

[@PatrickKennedy2]

Higher MTBF and look at the UBER too

[@bertnijhof5413]

I'm happy with my 1TB WD Black and my 500GB Seagate Barracuda, both have around 9 power-on years and the other SMART data still looks good :) I use both in my Ryzen desktop with three 500GB partitions, with a 2 x 500GB ZFS datapool (Raid-0) and a 500GB ZFS datapool at the end of the 1TB. The Raid-0 datapool has one dataset with copies=2 (Raid-1). To be honest they are supported by L2ARC/ZIL caches (90GB/5GB and 30GB/3GB) on a 128GB sata-SSD and my main stuff is run from a 512GB nvme-SSD. To compensate for the 9 years, I have two backups, one is on my laptop with a new 2TB HDD. My hobbies are VMs, I have all Windows Releases from 1986 to 2021 and all Ubuntu LTS Releases, the first 4.10, my first 5.04 and the last 22.04 Dev.Ed. The nvme-SSD contain the ~14 most used VMs. The striped datapool contains a dataset with the ~24 VMs still receiving updates and a Raid-1 dataset with my music; photos; family videos etc. The last datapool contains ancient archives of e.g 16-bit software and the frozen ~30 VMs :) Everything is lz4 compressed with a compression ratio of ~2. Since June 2019 my second backup is on a 2003 Pentium 4 HT (3.0GHz) with 1.21TB of HDDs, two 3.5" IDE (250+320GB) and two 2.5" SATA-1 (320+320GB). They have 2 to 4 power-on years and are powered on for 1 to 2 hours/week running 32-bits FreeBSD 13.0 on OpenZFS 2.0 :) :) The Ryzen 3 2200G desktop runs a minimal install of Ubuntu 21.10 on OpenZFS 2.0.

[@marcogenovesi8570]

Heyy, it is our boy Patrick back with an investigative article. I have read around that the rating is for an onboard 64gb flash cache (which is the same for all drives), not for the actual spinning platters of the drive.

[@ServeTheHomeVideo]

But then what about the 2TB drive?

[@jamesphillips2285]

@@ServeTheHomeVideo Why do you disbelieve that the 2TB drive would have the same cache?

[@paulstubbs7678]

It seems to be saying to me, this is a limitation of the mechanics, the actual underlying storage capacity has nothing to do with it, the limitation of being mechanical. Also, near the end of video tape, the mechanics got really cheap and crappy, a race to the bottom to get the last dollar - are hard drives on the same path?

[@mikebutler9332]

It's the same with the Seagate Ironwolf Pro's. All capacities from 20TB to 12TB list WRL of 300TB/yr, where as their Exos drives don't have one at all AFAIK. So I think the outlier here is WD putting a WRL on their Gold and Ultrastars.

[@annihilator9075]

I really want to know if it is a way to reject warranty claims not that they did something wrong with the drives.

[@Godzillaaaaa11]

@@annihilator9075 Seagate doesn't check workload for warranties, if the drive has failed and it's within the warranty period then they'll accept. Even their warranty statements don't mention checking against WRLs

[@cromefire_]

Can't wait until large flash drives are cheaper than HDDs... One day I hope I'll happen. Also you could (at least when I got my last drives) get Gold Drives for relatively close to the Red Plus prices (Amazon though, no bulk, I only needed 2 drives for my NAS) which did have a bit better rating, at least in MTBF (their load though).

[@theatermusic87]

If you do monthly parity checks in unraid or anything else that does full data read integrity checks... that only leaves you 2 drive writes worth of data before you are over the limit (assuming 14 days at rated speed to read/write) that sounds terrible

[@RJ_Cormac]

It's not just Western Digital, Seagate has the same ratings; IronWolf 20TB is 300TB/year, while the EXOS 20TB is 550TB/year.

[@ServeTheHomeVideo]

But Seagate has higher MTBF and such as well.

[@RJ_Cormac]

@@ServeTheHomeVideo that doesn't change the TB/year recommendation they will use for warranty claims the same as Western Digital. Go over that recommendation and risk your RMA denial.

[@turbo2ltr]

I've always bought WD drives. I haven't had issues with them, but they have pretty much eroded all my faith in them over the past couple years. I ended up unknowingly filling up my NAS with SDR WD Reds thinking I was getting good drives for NAS. Still pretty bitter about that one. At this point I think the Ironwolf drives are a better choice.

[@annihilator9075]

I get exos drives...they are cheap compared to the NAS versions, but I think it is because of the lack of warranty. (they have a warranty but is more difficult to take care of I think...never tried it)

[@alessandrozigliani2615]

As stated by Annihilator, Exos are good and actually not that expensive. I have had a lot of troubles recently with Ironwolf 4TB drives (bought 2019-2021). I've experienced weird problems from them, such as pending / uncorrectable writes and other S.M.A.R.T. errors that caused no data loss whatsoever (ZFS did not even detect there was a problem). And in fact, those problems were solved by a "repair" using SeaTools and have been working fine since then. But I had to replace them in my NAS, even though I could repurpose them. Still, over a batch of 9 drives that I bought at different times, 1 failed completely and 3 had this kind of problems. I am not so fond of ironwolf anymore... I have Seagate NAS 3TB drives that are still rocking after 6 years so I can definetely say that Seagate low end NAS drives has gotten worse in the last 5 years. There is no reason for it other than production quality going down because they are willingly making cheaper and less reliable drives.

[@drewstemen9597]

Just checked the specs for IronWolf Pro, and they're rated for the same 300TB as these are. I'm in the process of replacing a bunch of WD Red SMR drives with IronWolf Pro right now. 300TB/yr seems laughably low to me, but given that WD and Seagate are both doing it... *shrug* not much I can do about it.

[@nicktoale]

After getting stung by WD (Whirling Death) with 800 faulty drives failing in a load of video servers over a 6 week period in 2005 avoided them ever since. Not a fun 6 weeks trying to recover from that.

[@Daniel15au]

Something you seem to have missed (correct me if I'm wrong) is that the endurance for a HDD is yearly whereas the endurance for an SSD is usually across its whole lifetime.

[@ServeTheHomeVideo]

That was a big part of this video. SSD endurance is also usually quoted at 4K Random Write. Sequential writes mean you can get usually get more than 2x the rated SSD endurance.

[@dj-bn1fj]

For storage the real con you missed is rebuild time due to a failed drive. Example raid 5 the rebuild could be a week and what happens if another drive fail during rebuild? This is why storage companies like 4 and 6 TB drives so they can rebuild around 24 hours. Remember once a drive fails it's not long another drive will fail. I've had to rebuild storage after a 2 drive failure with all the data gone.

[@i_am_macgyver84]

I want to point out that WD Red Plus drives 8TB and larger are spec'd at 7200RPM. Sorry for the rant but I want to add. I've been using WD internal HDD's since around 1997, I have never had one problem with them. This past black friday I decided to buy 2 - 8tb red plus drives because of the sale price. I purchased them directly through WD. Upon receiving, I was in shock of how they were packaged. The one drive was coming up with bad / reallocated sectors and that number kept growing, the other drive vibrated so hard and was so loud, I'm thinking a warped platter??? Either way they are out for warranty, and even though I specified to the guy on the phone I want them packaged properly this time. I think I'm going to get the replacements packaged the same as the originals. Because of the SMR/CMR debacle I started my Unraid server with Seagate drives, then got a Toshiba, then I decided to buy 2 WD plus drives off Amazon which came in proper packaging, and all of those drives are functioning properly. I don't want to say I'll never buy WD again, but I will not buy directly through them ever again. The hassle wasn't worth the black friday sale price.

[@i_am_macgyver84]

1 last thing, they made me pay return shipping on their defective drives. Not counting my wasted time, they were still a good price, but now this is cutting into my sale price.

[@ahmadbas89]

this video should be 6-10 min max. the story telling style, I think, hurts your viewership. you can have a podcast where you can go into detail on how things came to be. keep the good work, you're definitely helping me through the journey of assembling my first home lab.

[@faithful451]

At around 12:00 the point doesn't really make sense. We're talking about the endurance of a single drive here, so in a simple mirror setup yes the total amount of data written is double (10TB x 2 drives) but it does not double or even substantially increase on a single drive - 10TB data is 10TB written on a single disk in a mirror setup.

[@WorBlux]

Also are we sure this is a warranty limitation, or is it just a performance guideline? The later makes more sense than the first. WD's warranty page just says 5 years for this drive, while it notes SSD drives may be limited by the endurance rating. Cache size varies accross the line size (512 or 256 MB) but the workload rating remains the same. It makes me think of this more as a performance rating/spindle or controller limitation. The datasheet does note that this can vary based on software and hardware configuration... This leads me to again think it's just a performance guideline that is rated fairly conservatively and accounts for moderately burst workloads and mixed read/write loads. Which leads me to think ZFS should be just fine on the drives, with the caveat a scrub will take 30 hrs or more to complete on an idle drive. This sort of sequential read is best case scenario though... At this size of drive the single spindle and lower RPM is a huge limitation. If you have a known idle/inactive window and pause/resume scrubs to keep in the window you'll be fine. Trying to mix scrubs with 24/7 data accesses/writes you'll likely run into underperformance issues.

[@DarksurfX]

Western digital’s sneaky move to SMR and then the split of plus and pro and the increase of failure rates all lead to me never purchasing WD REDs again. I’ve got a ton of old 4T WD RED, I was hooked on their reliability and raid capability. But now that seagate is stepping up to the plate with decent longevity after a ton of failures, I’ll be switching to Iron Wolfs. Already started that way.

[@zacker150]

Ironwolf pro also have 300TB workload rating.

[@noahlistgarten7832]

I'm going to be honest, I've had really bad luck with Ironwolf drives. I only run 13 of them, and I had 3 of them fail in as many months (1 each this past December, January, and February) , most of them just barely in-warranty

[@DarksurfX]

@@noahlistgarten7832 this is good to know. The seagate drives have started looking better based on backblaze stats. What size drive are you commonly seeing errors?

[@noahlistgarten7832]

@@DarksurfX Again, that's just anecdotal evidence. Yeah, Seagate stuff looks pretty good on BB's stats, but this has left a bad taste in my mouth. Hey, some people swear by Seagate and hate WD for the same reason as I hate Seagate, so I think how people feel is based on personal experiences more than actual data. I had a 6TB drive and two 8TB drives fail. Regular Ironwolf, not the Pros.

[@DarksurfX]

@@noahlistgarten7832 good to know. I recently had a few REDs die and had to RMA, then RMA the RMA, then after the warranty passed I just trashed them. Only the old REDs seem to keep going. These new ones aren’t as good as the old ones were. If you have the money hitachi drives are better (even though they’re owned by WD) I’ve even noticed some Toshiba drives are just rebranded hitachi with Toshiba firmware. They’ve been pretty reliable too.

[@maxwellsmart3156]

Seems Seagate Ironwolf Pro HDDs are rated for 300 TB/Y too, and they call that "built tough". Not a WD issue, it's an industry issue seemingly. You need to get a comment from the manufacturers because there are some issues that need to be addressed. Most users will not have an issue because they will use the drives without issue beyond the warranty period anyway.

[@berndeckenfels]

That is quite strange that the line which uses the supposed high capacity encoding won’t provide high capacity anymore. WAT?

[@AndrewHelgeCox]

If you pay for a higher endurance drive, do you get better technology or does the money go into a fund to pay out claims when there are failures inside the warranty period?

[@DocNo27]

I was wondering when we would hit the wall for what one could reasonably expect to pack onto a platter - how much was going to be too much. Well, it looks like it's starting to shake out finally :/

[@Liriq]

Could it be that this is due to a flash cache? Meaning that the whole line has the same type and size of flash cache, and the 300TB is the endurance rating of that flash.

[@ServeTheHomeVideo]

Even on the 2TB?

[@user-oo5ik7jn1n]

As the rating includes reading i might be a mechanical issue of the heads or head motors.

[@Liriq]

@@user-oo5ik7jn1n hm. Good point.

[@eDoc2020]

@@user-oo5ik7jn1n Disk reads could also be updating the cache.

[@keyboard_g]

WD has been a no-go for us since the SMR debacle.

[@ryan.crosby]

I wonder if these workload ratings from WD are actually based in reality, since they suspiciously haven't changed over many years, over different drive models, with vastly different capacities and read speeds. On a hardware level, what actually *is* the limiting factor for a read-only workload in a spinning hard drive? Is it head flight time over the disk? Because the heads are constantly flying over the disk regardless of whether the drive is reading from them or not. It seems strange to specify workload in generic TB/year when the heads are constantly in flight anyway. Any other metric would almost make more sense, like power-on hours, head loads/park cycles, spin-up/spin-down cycles, random seeks, temperature/hours, or literally anything else.

[@Chris173972]

I gave up with WD HDDs years ago. So many loops with sketchy crap forcing you to get the more premium range and in the past I've had alot of RMAs too. Seagate I have had for many years and had very few drives die. Bad to know they are still doing this rubbish.

[@ilovefunnyamv2nd]

here's some speak like a lawyer is in the room: Rated for UP TO 300TB. also means if the part fails within the first 5TB. it still met the rating. there is no minimum rating guaranteed. now that said, I haven't heard any complaints about the warranty not be honored if the part fails early, but I also don't dig through HDD forums unless I'm researching what to buy for the NAS.

[@AK-vx4dy]

In HDD wear don't depend on size, wear is mechanical, if there are more platters, wear is the same (excluding bearings of platters and motor). Also it's explains why reads are also taken in the account. This drive is imho for cheap backup storage with infrequent access and requires tunning of partol reads or scrub to long intervals.

[@jorelplay8738]

Except is not cheap

[@AK-vx4dy]

@@jorelplay8738 Cheap in $/TB meaning

[@bhume7535]

@@AK-vx4dy what does it matter if it last nowhere near as long as an SSD? if you have to buy multiple hard drives over the lifespan of an SSD then the SSD payed for itself.

[@AK-vx4dy]

@@bhume7535 if you only save backup and dont overwrite it (or rarely eg. once a month) then ssd don't pay themslef imho

[@bhume7535]

@@AK-vx4dy But that isn't the point of these drives. They're NAS drives, not cold storage and forget about them drives.

[@daninmanchester]

I guess it depends on your use case. 20Gb is great as I can get down to one drive (mirrored or replicated) and an SSD for a cache and you have small fast home NAS. Just don't scrub too often.

[@mhavock]

I have WD BLACK drives lasting in servers for 5-7 years, definitely impressed. The RED are less reliable, but mostly use in a RAID type environment, good for cheap temporary large storage arrays. Seagate seemed to have a lower reliability than the same level WD, so we never used them.

[@ibendover4817]

Wd blacks are definitely the most reliable hdds I've used. My oldest one is running over 8 years strong now.

[@spyrule]

I dont order any magnetic drives that arnt enterprise class anymore. The price difference is minimal, but the best warranty usually pays for itself. I also tend to prefer seagate drives over WD as of late.

[@samiraperi467]

10^14 b == 100 Tb == 12.5 TB. So you might have two (or more, probabilities are fun) incorrect bits on a 20TB drive.

[@Lollllllz]

That seems very low for a nas oriented drive if a 2016 256gb ssd has a 36.5TB endurance rating for the same warranty period.

[@tim3172]

How much is that SSD/TB compared to the HDD?

[@Lollllllz]

@@tim3172 not sure how much the 16nm intel pro 2500 retailed(mine came with a laptop) in 2016 but the samsung 950 pro offered a 10 year/200tb write endurance rating for the 256 gb unit for MSRP of $200 in 2014

[@friedrichschwarz1547]

Assuming the "endurance rating" is grounded in reality, it implies the culprit is either the movement of the moving parts wearing out, or the reading/writing head gets worn down through activity (but far less likely as higher capacity drives should distribute that load over the internal platters). still that low value suggests that these drives are not really meant for 24/7 usage, rather more like a desktop drive - mostly powered down.

[@Autotrope]

Awesome channel, wish I'd found it earlier. Saw it in a Jeff geerling video

[@cmuller1441]

It's the wonderful Optinand technology!

[@mntbighker]

I just bought the Seagate NAS drives (only 4TB). Both WD and Seagate seem to have a LOT of negative reviews on Amazon.

[@pianoplayer88key]

Several years ago Tech Report (I think it was them) did an endurance test where they hammered several SATA SSDs with continuous writes until they died. (Most were MLC drives but a couple were TLC, btw QLC didn't exist yet I think.) Most of them went way past their published endurance spec, with the longest lasting one being a ~250GB Samsung 840 Pro which lasted about 2.4 PB I think. I'd like to see someone repeat the same test, this time with both premium MLC/TLC NVMe SSDs (including enterprise ones with like tens of petabytes of endurance), as well as the basic budget QLC SSDs (with only a few hundred TB of endurance on a >1TB drive), and this time add hard drives to the mix as well - everything from the top-end datacenter drives (from Seagate, Toshiba AND Western Digital) all the way down to the most basic consumer drives, like the WD Green/Blue or Seagate Barracuda lines. Run the hard drives until they die - not just when they get bad sectors, but run them until they no longer spin up, or have the "click of death", or something like that. I wonder which would have more endurance - both in terms of total writes relative to their capacity, and in terms of how long they can stay continuously writing until they die? :)

[@mdd1963]

45Drives just did some tests where SSDs were dropping in performance after just 45 days of constant reads/writes....; data adds up quickly with sustained 500 MB/sec writes and reads...

[@jamesdk5417]

Thanks for the great info. That purple light on the side of your face was very distracting.

[@TomTravelling]

But ALL the HDD makers use the same calculation, not just WD. And are there NO other limits/rates on SSD usage, such as reads, age, radioactive decay?

[@vmoutsop]

Yeah but other NAS drives (Seagate Ironwolf & Toshiba N300) have 180TB workload ratings, what does that mean for the whole industry?

[@AdrianBacon]

This drive strikes me as an archival or backup drive. You fully load it once or twice a year, do a monthly integrity check and you’re fine. In zfs raidz configurations, it’s not actually a write amplification, but a workload rating amplification. In a 6 drive raidz2 vdev, a 1GB file is effectively broken up into 4 chunks plus two parity chunks of 250MB each for a total of 1500MB written, but the vdev has a combined workload rating of 1800TB (300TB x 6). Yes you’re writing 50% more data with raidz2, but the individual writes to each drive are 25% less than they would be for a single drive. This ratio gets even better as you go to larger vdevs with more drives. Think of it this way: do you want to write 1 TB to a 20TB drive with a 300TB workload, or would you rather write 1.5 TB to an array with a 1800 TB workload? This math doesn’t hold up if you load your array to full capacity, but you shouldn’t be doing that anyway. As your array goes above 50% capacity, you should start replacing disks with larger disks, or if larger disks aren’t available, offloading data to longer term archival storage, which again, I think WD is aiming their red series as that, home users or small businesses that need a handful of really large drives for archival or backup purposes where it gets fully loaded a couple times a year and scrubbed once a month but is otherwise very lightly used, if at all.

[@RealThore]

Looking up the prices I nearly fell out of my chair. The cheapest 20TB HDD (Seagate Exos) is about 450€ (including VAT), the WD Red is at 820€, who tf buys that stuff at such a prize?

[@mikefarino4368]

Enterprises where the density is far more valuable than the cost per TB

[@calvin_thefreak]

@@mikefarino4368 did you read? why buy wd if there are other manufacturers.

[@ikjadoon]

WD is grossly overpriced. Exos has many more authorized resellers and they are way better priced than any NAS drive once you look past 4 TB.

[@mikefarino4368]

@@calvin_thefreak my mistake, i thought you were referring to the massive drives in general, not specifically WD drives

[@JuxZeil]

Haven't trusted WD since the early 2000's as the HDD's they use in their portable units were very under par. I'll stick with the SG Exos drives for capacity.

[@samiraperi467]

I have some 4TB 3.5" WD drives in my NAS. The problem with the external enclosures where those were was that the USB adapter electronics tended to fail. I get you on Exos, they can be among the cheapest depending on price variations.

[@JuxZeil]

@@samiraperi467 I found that the HDD's were a problem too. Not only were they more expensive per GB, but they were all slower spin drives. I know that USB wouldn't need a faster seek, but still...the price. I also found a lot of times, the recovery of data when the disk was OK but the converter bridge was dead...well ...non existent because they formatted the drive with a propitiatory standard....couldn't be read without the original hub DRM key...even if you set the drive to anyone in it's properties. 😒

[@asabhcir3636]

I have two 6TB WD Red Plus in 2 Bay WD NAS in Raid 1 since 2016. Works great, no issues, so far so good.

[@JoshFisher567]

I have gone with western digital almost my entire life. I've had issues with Seagate, in particular DOA drives. Now,. I work in IT and a lot of my friends work in IT and it seems. Like everyone has had major issues with one or the other. Some friends won't touch WD, I would typically not go with Seagate. I know the last 2, maybe 3 Seagate ordered where DOA. Bad luck, yes,, but when you return the 3rd DOA. And get a WD that has been going since 2011. It's not in use but it works and has over a petabyte of data written according to Crysta Disk Mark. Obviously no important data is on that drive. There are people who can say the same about Western Digital. It only takes 2 or 3 issues or bad experiences with one brand before moving to another. I do think WD buying SanDisk will help them with SSD's going forward also but I could be wrong. I've personally had better luck with WD but hard drives (spinning) tend to either die early, or last day at least 2 years in my experience.. Every hard drive (and electronics) manufacturers just wants the product to work until the warranty is up. Sometimes you get lucky, others times you don't. Every spinning drive I have that has lasted 5+ years is WD, but I have only bought WD drives for the most part so of course they are

[@diavuno3835]

I feel like Western Digital makes their product lineup complicated

[@mightylink65]

WD used to be known for high endurance drives... did they pull a bait and switch? Or was it more like a long con? It's like going out and buying a well known high quality brand like Bentley and finding out they've been producing lemons for the last few years...

[@ericneo2]

They sold to seagate 🤦‍♀

[@thegeforce6625]

Or low yields?

[@tianmul8134]

I checked the Ironwolf Pro drives from Seagate, which should be WD Red Pro competitors, and they have the same 300TB/year workload rating for 20TB drives. I think this is more about hard drives being crap in general instead of WD Red Pro being particularly bad. This doesn't make the situation any better though. Even when you only have a 6TB drive, 300TB/y for 5 years is about 250TB workload for 1TB of storage, which is about the same as consumer QLC drives. I always thought that hard drives are suitable for NAS partly because they don't have endurance issues, but apparently, that's not the case at all.

[@bluegizmo1983]

WD is just way too shady of a company these days. They burned me with that whole "WD Red drives are CMR!" Lie, and that was the last straw for me. I'll never use another WD drive ever again.

[@TheJensss]

I bought 6x of the Corsair MP510 1TB drives that has TBW rating of 1700TB. What is the reason this drive has such high datawrite rating? I have already written 300TB on them and they are still working as new.

[@ahothabeth]

Note the Corsair MP510b have a far lower TBW @ 720TBW. I was going to but a couple recently and found out about the MP510b, because not all seller are clearly denoting the difference I stayed my hand. MP510 CSSD-F960GBMP510B 960GB 720 TBW

[@TheJensss]

@@ahothabeth Yeah, I remember that was a little "scandal" within the tech community when it was discovered. I was lucky and bought my drives about a year or two before this was relevant, so i'm sure I got the 1700TBW version. Do you know what is the difference between thees models?

[@klightspeed]

Their workload rating would suggest that they expect a scrub to be performed no more than once a month, versus the usual recommended weekly scrubs. If they included extended SMART self-tests, then the default smartd weekly extended self-test would all by itself exceed the workload rating on anything larger than a 3TB Red Plus or 4TB Red Pro drive. My WD40EFRX-68N drives at home are sitting at about 210-230TB/year just from weekly RAID6 scrubs (the older EFRX-68W drives don't support the Device Statistics SMART page)

[@agnag1]

If you do a full scrub every 3,5 weeks you almost exceed their yearly rating, doing nothing else (20TB capacity | 300TB/yr drive)

[@LucasHolt]

I'd like clarity on exactly what this extremely low workload rate is. Obviously, it gets them out of needing to warranty drives. Are they that unreliable that we shouldn't even buy them? Do they have a high rate of error compared to older drives? With a SSD, it's clear what happens when the drive wears out. It's less clear what happens if you go way past the workload rate on hard drives.

[@Vatharian]

Waaaaaaait, the moment when you said it loud, it dawned on me that it has endurance rating of 15 drive writes per year. I was thinking before that's not a big deal, but when you compared to low capacity QLC drive, it's much lower. It's 0.04 DWPD. Holy shmockey, that hits hard. Edit: I just realized it includes reads. Oh.

[@mattatwar]

The Ironwolf pros show 300TB/year for the 20TB drive

[@ServeTheHomeVideo]

You are correct, but IronWolf Pro's have slightly better reliability specs outside of the workload rating. I also wanted to point out that the 20TB models, with OptiNAND, are likely the same as the Gold/ HC560 mechanically, just with different labels. So the low reliability specs are most likely entirely marketing.

[@jakeAU44444]

yeah on a 500gb ssd 970evo that has just over 800TBs written and still going strong.

[@XtianApi]

Got 4 of these delivered today. One failed out. Loud noises, Nas failed it. Great. Spinners aren't as good as ssds. We know this. But you can't get a 20tb ssd for the same price. Makes sense. Real enterprise stuff that I use at work are thousands of dollars per drive. Too rich for my blood.

[@SnoJMusic]

how would you describe the noises? I've got an 18tb that squeals/whines occasionally, been doing it since new and I've put 1000s of hours of intensive read/writes and if anything it's started to make the noise less often. I sorta want to get a couple years out of it before I warranty claim it haha

[@XtianApi]

@@SnoJMusic mine was dead drive noises. It was like if the normal read write sounds were way louder and repeating the same patterns forever until you turn the drive off. Your sound is interesting. So it's the only drive like this? Maybe a squeely bearing?

[@sopota6469]

Western Digital: Destroying their reputation one product at a time. I loved sooo much the original WD Reds and Blacks. After the SMR snafu I never installed one of their drives again.

[@galen__]

Monthly verification reads of 10TB+ data per 20TB drive is going to eat into that 300TB/year R/W endurance 🤦‍♂️

[@shinigamilee5915]

Worked for WD developing drives both spinning and solid and I can assure you that they only have one goal. Make money. They have actively stifled innovative products and spent billions of dollars to stop SSDs from becoming cheaper.

[@dumboy886]

On the one hand, WD seems to be getting in the news recently for bad practices, on the other hand, when I look at Seagate drives failure rates on backblaze reports, I'm really concerned. Ugh.

[@deth3021]

Wonder if the limitation is the write heads, the controller, or just further reducing the warranty .

[@jms019]

So you can spool scrub about twenty times ?

[@mattw3406]

Patrick - have you done any recent testing on the WD external white labeled "schucked" drives? They seem to be Ultrastar 560 data center drives (14tb variants)

[@ServeTheHomeVideo]

We have a bunch of 18TB shucks

[@icebalm]

A 20TB drive with an unrecoverable error rate of more than 0 in ~12TB of reads meant for NAS use is completely worthless. You're never going to be able to rebuild an array if one fails since you'll have to read all of the surviving drives to do it at which point, according to the specs, you're virtually guaranteed a read error.

[@nathanmiddleton1478]

Not being a common consumer of this kind of drive, I'd just go buy 'the better' in this kind of category without reading the fine print and probably get caught in a trap like that. "Read AND write" is a little insane given the size. As you pointed out you could easily write to a flash drive, even a consumer flash, and read from it without much consequence until the cows come home but with this be careful because you might void your warranty? Worse yet, is there hardware inside the drive tracking the usage so that any type of RMA situation could be check on the WD side and be held against the user for some kind of small infraction like your given example?

[@DergEnterprises]

Maybe the hard drives are at their technical limit and WD is doing their best to increase capacity. MFM/RLL hard drives and later IDE hard drives had limits that couldn't be overcome.

[@zaz5190]

Recall many years ago, when cellphones were first introduced to the market, users were charged both making a recall and answering a call. 2022, WD is counting the HDD reading, really???

[@gravesclay]

Pure and simple... Market segmentation. Red "pros" are just that, soho, "pro" ie NOT ENTERPRISE. They are segmenting the market intentionally. A home user seeking a premium product will go for this. If you need enterprise level performance and assurances, buy the enterprise product. The exact thing could be said of any other tech product, enterprises buy "pro" model laptops because the drivers are standardized and you can get better service on things like warranty issues, as opposed to consumer grade products that are throw away. Put another way, sure a desktop can be used as a server... A really expensive desktop may be even a decent server, but in the enterprise we are buying rackable units with XEON or EPYC cpu's stripped down motherboard feature sets, extra expensive add-on cards etc in the name of stability. Will they outperform the desktop configured similarly spec wise but with the consumer level parts? Maybe, Maybe not, but I'm not risking my critical services on finding out.

[@ServeTheHomeVideo]

The only challenge I have with that is that the Pro line was the >8 drive line, specifically for up to 24 HDDs in a chassis. At 24x 20TB that is almost half a PB of raw storage per chassis which is not exactly a low-end box. 1PB just a few years ago was a very high-end NetApp.

[@craigshea2930]

@@ServeTheHomeVideo But if you're going to store that much data, then why even bother with the Red Pro at all? Just go Gold. They're also meant for huge amounts of drives stored in a chassis. I've NEVER seen a Red Pro priced for less than the same sized Gold. Even if the Red Pro were priced less than the Gold, it would have to be significantly less expensive, or else it wouldn't be worth it when you take into account the Gold's better Warranty (5yr vs 3yr), MTBF (2.5M Hrs vs 1M or 1.5M Hrs) and WRL. At the current (and recent historical) Red Pro prices, I just don't see how I could justify recommending to anybody the Red Pro line of drives--even if they only needed 2! (Based on what I've read in the comments on Seagate IronWolf, I'd say the same and just recommend the Exos--although, to SeaGate's credit, the IronWolf is priced lower than the Exos drive! Which is a huge difference between Seagate and WD!)

[@thedopplereffect00]

Isn't the thing that actually wears out the head clicking back and forth? Why not just measure that number directly and warranty based on that? That way you could test your workload yourself to see how bad it would be.

[@nickbarkas5774]

It's not funny on the terabyte era, the workload rating has stayed the same while capacity kept going up. Are they going to produce disks that will get out of warranty once they get written 10 times or less?