It allows the the media companies access to the failed hardware to do their own autopsy on it, and it saves the users from needing to go through a painful RMA process, complicated by companies not willing to admit fault.
The only way to know if anything even works to begin with is to read all (poorly written) manuals front to back taking notes, then procure the rest of the parts and rigorously test them yourself within all of their 30 day return windows. And even then you're virtually guaranteed to miss some glaring issue.
Just last week, an obscure forum post from someone who already went through the tech support/RMA gamut saved me from wasting a month + $5K on a build with a motherboard that doesn't support sleep mode, which the manufacturer ASRock doesn't mention anywhere.
Ymmv
Eg the ADATA XPG SX8200. Look for whole drive fill speed benchmarks, if they use the whole drive as cache, the first third is fast (usually the SLC area is much smaller).
Linus tech tips covered this issue before: https://youtu.be/K07sEM6y4Uc
The equivalent is in short "look at what the enterprise storage vendors are putting in their arrays, at 10x markup. (No, shiny things like Pure/Rubrik/Cohesity are not enterprise storage).
It all depends what you're putting on things. If you buy 5 drives from Newegg for your house and double-parity them or do zfs checksums, etc., you're going to have a bad time when a bunch fail at around the same time because it's an issue with the drive. Yet you do kinda want all the same/alike drives, because the stripe is only as fast as the slowest drive.
So look at what all the vendors picked after they tested the crap out thousands of them all. Me, I personally just mirror everything between two machines with different brand drives, and hope they won't fail at the same time. Once a year I dump an image of everything on an offline big-ass drive - the cheapest spinning big rust that I can buy - and call that my "airgapped vault."
Industry marketing (and accompanying irrational pricing) has basically persuaded consumers to choose an inferior product.
https://www.anandtech.com/show/18863/micron-updates-data-cen...
Either way, that's great news - and the ~$0.32/GB they mention (only $600 for 1.92TB!?) for Micron SLC is absolutely amazing value, if you consider that this other article I submitted not long ago mentions ultra-cheap TLC SSDs with NAND from an unknown manufacturer costing $0.10/GB (I even have a comment there lamenting the lack of logically-priced $0.30/GB SLC SSDs!): https://news.ycombinator.com/item?id=35382252
If SLC NAND went extinct, that's because both the sellers and the buyers (read: customers, aka end users) didn't see value in reliability as much as other factors like storage density and price-per-bit.
You, as someone who does want reliability above all else, are an outlier.
Many PC enthusiasts also want reliability.
The marketing might include a x-million r/w cycles in it, but it's going to be way under presented vs the speed.
That was years before this issue but my rule of never buying a SanDisk product again is serving me well.
As (sometimes six-year-old) Samsung cards get retired, I've gone to... Samsung. This time, I'm getting their cards intended for surveillance cameras and other write-intensive duty.
Switched to Sandisk then it has not failed after a year, I probably will never buy Samsung cards.
I have not bought a WD or SanDisk drive since, I'm still very pissed that I spent days debugging this issue, decided I needed to scrap the entire machine and then still had the issue. Who thinks of a bad drive as a reason you can't even boot into BIOS?!
https://www.anandtech.com/show/15673/dell-hpe-updates-for-40...
this one was a real pain in the ass to deal with. always make testable backups, people. backups are not the same as redundancy. a 1, 2, 3, even 7 or 14 day recovery point is far better than poof it's gone.
I know there are tools like Restic that can do what Crashplan did, but what's the TrueImage equivalent for Linux? Ie something I can use to clone my primary disk nightly in the background, including boot partition, incrementally with periodic full clones, and that supports resizing partitions (both up and down) in case the replacement disk isn't of equal size?
I know of Clonezilla but from their front page it can't do incremental which is a showstopper. With TrueImage each incremental image takes only about 2-3GB per day on average, full is 500GB. It also seems to only support resizing up partitions which isn't great as it means I can't easily use old disks as emergency restore targets like I did when my OCZ died.
I know ZFS root + send/receive is an option but as much as I like ZFS, I'm not comfortable running it on root yet.
OTOH, enterprise parts are built and supported towards conservatism and reliability.
There is crossover and a spectrum between the 2, but this case isn't a complete surprise.
Storage (from well known brands) used to be the most reliable component. Not sure what is going on, but feels like quality control is not as good.
https://community.wd.com/t/wd-blue-250-gb-sa510-issues/27981...
https://community.wd.com/t/wd-blue-sa510-sata-ssds-critical-...
Actually, I take that back. Maybe solidigm/SKHynix?
If you actually want a durable SSD, your best bet is probably to buy "new old stock" of a model that is 1-3 years old and has proven to be reliable.
Of course, it's not always easy to gauge the reliability of a product based on forum posts. A popular product that is 99.9% reliable will... still have thousands of unhappy people crying on forums if they sell millions of units.
In the end, you should probably just have some kind of daily (or better) backup system so that it's not a huge deal if your drive kicks the bucket. That probably makes more sense than obsessing over reliability in a world in which we as consumers don't have much insight into actual failure rates.
link: https://www.amazon.com/varencs-4tb-ssd-rec-ps-did-you-know-y...
They use them as a boot/system drives without a big load. And when you have more than 500 drives some would die just by pure luck (or lack of it). Mishandling, static charge etc...
> 60DWPD for 5 years is basically 100K endurance, the same as true SLC.
Cool, did not make that realization, thanks!
Batteries and fans, I guess.
Everything else doesn't really fail unless the user does something silly (toasting it via overclocking/overvolting) or a capacitor leaks.
Spinning-rust HDDs used to fail a lot because they were well, moving parts with incredibly small tolerances and computers get banged around sometimes.
It's kind of funny that SSDs carried on the tradition of failure despite not having moving parts.
I may end up buying SD cards at the local farm and game store for the trail cameras because hopefully nobody is counterfeiting those.
Why not? Company needed SD cards, they bought them and packaged with their kits. It could be from literally any source. Even with official SandIsk partnership.
- Gets it out early for third-party verification such as these YouTube channels that are doing testing.
- Keeps it transparent that they are still working on the issue.
- Allows them to experiment with more exotic BIOS (in the general case).
Maybe what is missing is some sort of message like "This is an early preview, we are still doing internal verification before we can provide a stable release with a fully warranty"?
Also IIRC the phrasing was "installing this will void your warranty" which is not true. It would have been better if they were clear that "damage done to your hardware by non-stable BIOS are not covered by warranty".
The SanDisk is roughly the size of a credit card, maybe 1 x 5 x 10 cm
https://www.youtube.com/watch?v=uDT_wl2sPF4 (1 min teardown)
Speaking of firmware, there appears to have been an effort at making an open-source SSD, but it seems rather inactive: https://news.ycombinator.com/item?id=19629298
They realised it's easier to keep making a profit when drives keep "wearing out" (i.e. failing to be a data storage device) on a consistent and short(ening) schedule. Just like SLC, Optane was too good.
"small" is relative. 8TB of QLC is 2TB of SLC. They will both cost the same (if anything, the SLC might even be cheaper from a firmware/controller development perspective) yet the former might last a few years, and the latter several decades.
And in reality, the kind of consumer who would spend $1k on an SSD is going to move on from it within two or three years anyways in favor of a newer drive with a faster interface.
...or expect that it will last much longer than a cheaper one.