Mailinglist Archive: opensuse (686 mails)

< Previous Next >
Re: [opensuse] SSD and smartctl: Percentage Used Endurance Indicator: 34 ?
\

I do have /var with leafnode/usenet (lot of small files written) on that
SSD
but what I read on the net this shouldn't be a problem and approx. 2
years
is not that much of lifetime. If that all is correct it would mean, that
after another 4 years I have to exchange the SSD.


Yep.

~6 years is that the normal lifespan of SSDs nowadays?


It is not based in time of use, but in actual usage. Flash media,
including SSD, have a limited number of write operation allowed. After
that, they are no longer usable.


Ok, so with the current average usage there will only be something like
another 4 years.
Good to know.

Are there some possibilies that smartctl may warn me at
-say- "90~ Percentage Used Endurance Indicator" without making my own cron
job and some script?

Your implication is that when it hits 100% it's lights out.

That's not how I understand it. Your SSD should be overprovisioned.10
or 20% to allow the garbage collector to efficiently maintain a
collection of pre-erased erase blocks for your writes to go to. The
goal of the wear-leveling algorithm and the garbage collector is to
keep the write cycles fairliy evenly spread across your EBs (erase
blocks), but there will be variation across them with some of them
likely still having the original data on them even near the end of
life for some of the EBs.

When the garbage collector starts to see some EBs hitting near end of
life, it should start swapping them out with some of the ones that
have extra stable data on them. By doing that a new supply of your
EBs is added into the mix. Thus when the most used EBs are hitting
near EOL, a new supply of relative new EBs might suddenly be swapped
in.

The end result is that it is very hard to know when the true EOL for
the SSD is. Further, as individual EBs do actually hit EOL, that just
pulls them out of the available pull for wear-leveling / garbage
collection. The first ones to go won't have a noticeable effect. As
more and more are pulled out of the rotation, the SSD performance will
start to drop.

Discarding your unused data blocks from time to time will maintain
performance even as some EBs hit EOL. Ext4 supports fstrim to do
that. Adding fstrim to a cron entry once a month (or every reboot)
might help keep your SSD in better shape.

I don't know what will happen when your SSD hits 100% used, but it may
just be that SSD starts to see a gradual drop in performance.
Further, a backup, mkfs discard, restore of all the data on the SSD
would allow the wear leveling algorithm to get a full refresh of all
the EBs. You might want to consider doing that now so your SSD can
get a fresh start at the wear leveling.

Greg
--
To unsubscribe, e-mail: opensuse+unsubscribe@xxxxxxxxxxxx
To contact the owner, e-mail: opensuse+owner@xxxxxxxxxxxx

< Previous Next >