• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

SSD Hardware life

E

Epyon

[H]ard|Gawd
Joined
Oct 25, 2001
Messages
1,514
So I have a Nas full of 4TB nvme drives. My question is when i go offshore for 4 months and come home for 2 months to do my work or what ever How long does the hardware last? I am thinking about getting some 8TB drives in the nas but if the hardware will just rot out in 5 years.
 
Unless you have issues with moisture or something, I very much doubt that anything will "rot out" in a few years.
The more likely concern with drives wearing out is usage. What kinds of drives are they (TLC vs QLC, etc.) and how often are you writing to them?
 

Crucial P3 4TB PCIe Gen3 3D NAND NVMe M.2 SSD, up to 3500MB/s - CT4000P3SSD8​


As for writs.Not very often its more for fast storage when i need to render something but not being used for 4 months at a time it is good to hear they will last a few years. i guess more in my case?
 
Trying to understand rot out. They will be fine just sitting in the NAS.
 
it will be fine. power them up and use them as needed. the lack of use might actually extend their life
 
Some relevant stats for Crucial P3:
  • "Rated at MTTF greater than 1.5 million hours"
  • Endurance rating: 800 TBW (that's relatively low, but obviously not a problem if you're not actually writing that much to them)
  • They're QLC drives, so less durable than TLC
 
I think the OP was thinking of bit rot when NVMe/SSD devices are powered-off for extended periods of time. I still haven't validated any such claims in the wild yet.
 
djstarfox said:
I think the OP was thinking of bit rot when NVMe/SSD devices are powered-off for extended periods of time. I still haven't validated any such claims in the wild yet.
Click to expand...
It probably does take several years at least. It's just a voltage level, which will eventually degrade to the point where the controller will view a NAND cell as a 0 instead of the original 1. The more states per cell (QLC+) would probably make that happen faster.

EDIT - apparently it's 10+ years unpowered for SLC, 3-5 years for MLC, 1-3 years for TLC, and 1-2 years for QLC https://storedbits.com/how-do-ssds-retain-data-without-power/
 
Last edited:
I have a Samsung 980 1TB NVME and I wrote almost 1PB to it in a week and it's at 55% life.
That is like writing the entire capacity of the drive 1000 times.
Under normal use I think the hardware would become obsolete before it runs out of life.

CrystalDiskInfo_Samsung 980 Pro 1TB 55 percent good.png
 
Ok because I was googling the wrong thing I was looking for the physical hardware itself that might degrade over a few years.
 
I built a server in 2013 and it has run 24/7 since then, with two Samsung 830 SSDs mirrored for the OS and they’re still going strong. As a matter-of-fact, I think I’ve only had to replace one fan and one spinner in the entire time.
 
chameleoneel said:
Overall, SSD's have not proven to be more reliable than Platter Hard Drives.
Click to expand...
If you run 1000 SSD and 1000 spinners for a year, you'll have a pretty good idea of which fails more often. From my experience, it's the spinners, SSDs also fail, but not as often.
 
chameleoneel said:
Overall, SSD's have not proven to be more reliable than Platter Hard Drives.
Click to expand...

Pros and cons. One of those situations where vague generalizations probably aren't helpful. The most reliable drive is going to be the one that is the best match for your use-case scenario.
 
LukeTbk said:
Seem to be a really large gap at least for heavy usage and mid-term life.
Click to expand...

To be totally fair, I think Backblaze uses SSDs for their OS and spinners for user storage, so it's not apples to apples by use case. Wild speculation follows. But I suspect both kinds of drives are going to be mostly idle in their system. User backups are mostly going to be write once, read never (+/- consistency checks), and there shouldn't be that much storage churn once a new system gets mostly full.

If you're doing a whole lot of writing, you can wear out an SSD early, but a spinner wouldn't be write speed limited, so if that's your use case, SSD vs spinner reliability doesn't matter as much as which can do the job.
 
I think they compared ssd vs hdd both used as boot drive here as well, at least in the table at the right side:
But one place we have both HDDs and SSDs is as boot drives for our storage servers.
If we look at the lifetime results for the HDD and SSD boot drives, the difference in AFR is less, but still significant.
What we have are two groups of drives, one HDDs and the other SSDs, which have performed the same functions in the same environment over time

But maybe that a use case that advantage SSD but blackblaze boot drive log a lot vs a regular os drive.
https://www.backblaze.com/blog/backblaze-hard-drive-stats-q1-2021/
 
Last edited:
1-2 years of data retention for TLC and QLC (the overwhelming vast majority of drives in use)??? That’s awful.

As Eric Cartman famously said, “Put a disk in it, and make it stay”
 
You must log in or register to reply here.
Back
Top