SSD Lifecycle

[Infrasonic]

https://blocksandfiles.com/2021/10/01/b ... sk-drives/

Cloud backup and data storage provider BackBlaze is finding its SSDs fail at nearly the same rate as its disk drives at the equivalent stage in their life cycle.

A Backblaze blog by Andy Klein, its oddly titled Principal Cloud Story Teller, outlines how Backblaze used to boot its systems off disk drives and then started using SSDs instead. It monitors its disk drive reliability and does the same for its SSDs, so it can compare their failure rates. To make the comparison fair, it compared the SSD failure rates to the HDDs used for boot drive duty at a similar age in their life cycle and found similar failure rates.

Cont.

In other good news I see the price of silicon has gone up 300% in the past couple of months as a knock on effect of energy prices causing the Chinese to reduce production - so I can't see component shortages being resolved any time soon and OEM's swapping parts out for cheaper alternatives becomes even more likely. Caveat emptor!

[xeny]

That may be a pretty atypical workload though. I don't know about the details of how BB use their boot drives, but most server boot drives see a flurry of reads at boot, and then just writes to log files. That's a pretty easy life for a drive, mechanical or SSD.

Boot drives on desktops and laptops used interactively see much more in the way of ongoing random I/O with more frequent reboots, application launches, web browser cache read/writes etc. I can imagine that is a much tougher environment for a mechanical drive (lots of seeks especially). I've no idea how the change in workload is likely to impact an SSD.

[Infrasonic]

That may be a pretty atypical workload though. I don't know about the details of how BB use their boot drives, but most server boot drives see a flurry of reads at boot, and then just writes to log files. That's a pretty easy life for a drive, mechanical or SSD.

Boot drives on desktops and laptops used interactively see much more in the way of ongoing random I/O with more frequent reboots, application launches, web browser cache read/writes etc. I can imagine that is a much tougher environment for a mechanical drive (lots of seeks especially). I've no idea how the change in workload is likely to impact an SSD.

True, it's writes that tend to kill SSD's.
Servers tend to be on 24/7 as well - but boots take much longer - minutes in many cases.

What would be more interesting and relevant is if they have any SSD/HDD storage or cache comparisons - I note quite a few mid range SOHO and up NAS' (Synology et al) now use dual NVMe SSD cache to speed writes up - and you can get high write NAS grade SSD's (with the corresponding hike in price...).

[Infrasonic]

From the Backblaze blog..https://www.backblaze.com/blog/are-ssds ... rd-drives/

Comparing Apples to Apples
In the Backblaze data centers, we use both SSDs and HDDs as boot drives in our storage servers. In our case, describing these drives as boot drives is a misnomer as boot drives are also used to store log files for system access, diagnostics, and more. In other words, these boot drives are regularly reading, writing, and deleting files in addition to their named function of booting a server at startup.

So maybe more similar to a system C drive than initially presumed then.

[Breelander]

From the Backblaze blog....

I do like Backblaze's blogs. This is one of my favourites that I often refer others to...

What SMART Stats Tell Us About Hard Drives

https://www.backblaze.com/blog/what-sma ... -failures/

For SSD endurance (and an entertaining read) try this.

The SSD Endurance Experiment: They’re all dead
I never thought this whole tech journalism gig would turn me into a mass murderer. Yet here I am, with the blood of six SSDs on my hands...

https://techreport.com/review/27909/the ... -all-dead/

[Infrasonic]

One element none of the articles and blogs ever mention is stable power.

I can't prove it but I have a suspicion the one SSD failure I suffered was more to do with the constant brown/black outs in my area as Electricity North West were constantly fiddling with the local infrastructure to supply the unending stream of new flats being built.
It certainly wasn't high writes.

Enterprise kit has ECC RAM with battery back up, smoothing capacitors et al in much of its critical hardware (in addition to beefy in line UPS/backup generators) and if you are prepared to pay for the more expensive SOHO NAS SSD's you can get smoothing caps there too.

There's some good DIY videos on YouTube that show ways of doing cheap home equivalents of some of the above - avoiding actually modding motherboards with the attendant warranty voiding issues.

[88V8]

One element none of the articles and blogs ever mention is stable power.
I can't prove it but I have a suspicion the one SSD failure I suffered was more to do with the constant brown/black outs in my area as Electricity North West were constantly fiddling with the local infrastructure to supply the unending stream of new flats being built.

That's interesting. I run the PC through a UPS. Often more than 240V in, but 220V out.
Perhaps that will help.

V8

[Breelander]

I can't prove it but I have a suspicion the one SSD failure I suffered was more to do with the constant brown/black outs in my area....

Not an issue here - all my machines are laptops