Disk I/O as a bottleneck?

[shimi]

On Sun, May 8, 2011 at 1:21 PM, guy keren <choo at actcom.co.il> wrote:

> On Sun, 2011-05-08 at 12:26 +0300, shimi wrote:
> > On Sun, May 8, 2011 at 12:01 PM, guy keren <choo at actcom.co.il> wrote:
> >
> >         do you have the ability to extract wear leveling information
> >         from your
> >         SSD? it would be interesting to know whether the drive is
> >         being used in
> >         a manner that will indeed comply with the life-time expentency
> >         it is
> >         sold with (5 years?), or better, or worse.
> >
> >
> > I don't know, how do you extract such information?
> >
> > The rated MTBF of my specific drive is 2 million hours. If I still
> > know my math, that's some 228 years....
> >
> > -- Shimi
>
> wear leveling has nothing to do with MTBF. once you write ~100,000 times
> to a single cell in the SSD - it's dead. due to the wear leveling
> methods of the SSD - this will happen once you write ~100,000 times to
> all cell groups on the SSD - assuming the wear-leveling algorithm of the
> SSD is implemented without glitches.
>
>
I know... I was referring more to the "life time expectancy it is sold with"
when I quoted that number. Unless block write fails do not consist of "a
failure", and if not, I don't know what is, in an SSD :).

Obviously on a DB data disk it is going to happen much faster than on my
desktop.

b.t.w. IIRC when a cell dies, it does so "gracefully"; I.e. no data is lost,
and there are spare blocks for that case... and even when they're all full,
you just get to the point that you still have your data read-only. I vaguely
remember I read that somewhere... and if it's indeed like that, this is
still way better than a regular hard drive - those tend to usually take all
your data with them, and are much more sensitive to many things (shock -
physical/electric, heat, etc...)


> note that these writes don't come only from the host - many of them are
> generated internally by the SSD, due to its wear-leveling algorithms. an
> SSD could perform several writes for each host-initiated write operation
> on average. intel claims their X25-E has very impressive algorithms in
> this regard. it'll be interesting to check these claims with the actual
> state of your SSD.
>
> I know that too :)

>
> fetching wear-leveling info is SSD-dependent. you'll need to check if
> intel provides a tool to do that on linux, for your SSD.
>
> Google didn't help me a lot in that regard (I understand there's a Win$ SW
maybe to do that; "Unfortunately", I don't have code from Redmond anywhere
inside my chassis.)

So if you find something yourself, I'm willing to test...

-- Shimi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cs.huji.ac.il/pipermail/linux-il/attachments/20110508/b08081b8/attachment-0001.html>