-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

James G. Sack (jim) wrote:
> I also read the pdf from google, and found it both very readable and
> very interesting. The lack of confirmation of traditional guidelines on
>  temperature and activity (after infant-mortality effects are weaned
> out) is especially surprising. The inadequacy of smart data as a model
> for predicting failure is almost as notable.

I think they overstated the inadequacy of smart data. It was useful in
predicting half of the failures. That is a lot better than nothing.
Hopefully the people who developed SMART can take a look at these
results and figure out what to monitor to catch the other half. I
suspect they are sudden catastrophic type failures which you cannot
easily monitor for.

> I glanced at the second paper and it looks harder to get into (is that
> just me?).

It does go a bit further into the statistical methods but I got into it
just fine.

> One of the points they didn't seem to grasp (or feel worth emphasizing)
> was that returns testing "no problem" do not mean there _was_ no
> problem. 

I thought they did explicitly address that issue. Section 2.3 of the
google paper addresses this very point. 2.1 of the Schroeder paper
addresses this also. They both seem to define failure as "anything which
causes the end-user to replace the drive". They might send the drives
back to the factory and the factory finds that they are good but they
are still failed for the purposes of the study.

> Anyway, IT can probably throttle back the A/C, eh?

Of all of their conclusions I am most skeptical of the heat related
conclusion.

- --
Tracy R Reed              http://ultraviolet.org
D4A8 4860 535C ABF8 BA97  25A6 F4F2 1829 9615 02AD
Non-GPG signed mail gets read only if I can find it among the spam.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iD8DBQFF3Amr9PIYKZYVAq0RAisSAJ97Hq8Rf5vAvkNtgS1VMH1X+c/i4ACeICQf
K+g7tdF+KJ3YFBHoMKB6rRo=
=GgoV
-----END PGP SIGNATURE-----


-- 
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list

Reply via email to