Date: Tue, 19 Apr 2016 16:45:13 -0500 (CDT) From: "Valeri Galtsev" <galtsev@kicp.uchicago.edu> To: "Jon Radel" <jon@radel.com> Cc: galtsev@kicp.uchicago.edu, freebsd-questions@freebsd.org Subject: Re: Raid 1+0 Message-ID: <26823.128.135.52.6.1461102313.squirrel@cosmo.uchicago.edu> In-Reply-To: <5716A422.6000707@radel.com> References: <571533F4.8040406@bananmonarki.se> <57153E6B.6090200@gmail.com> <20160418210257.GB86917@neutralgood.org> <64031.128.135.52.6.1461017122.squirrel@cosmo.uchicago.edu> <441t61efpv.fsf@be-well.ilk.org> <13260.128.135.52.6.1461084740.squirrel@cosmo.uchicago.edu> <57166A7F.2030104@radel.com> <44929.128.135.52.6.1461091476.squirrel@cosmo.uchicago.edu> <5716A422.6000707@radel.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, April 19, 2016 4:33 pm, Jon Radel wrote: > On 4/19/16 2:44 PM, Valeri Galtsev wrote: >> My opinion that your conclusion that the drives having the same MTBF >> (or similar lifespans) are likely to die simultaneously contradicts my >> understanding of probability theory (or statistics basics). People >> life expectancy in some country is 75 years. Likelyhood that two >> _given_ people will die at age 40 is negligible. Even twin brothers >> are unlikely to die during the same week or Month. This one is between >> fundamental assumptions statistics is based on. Just my humble >> opinion. Valeri > You're confusing presence of correlation with a layman's "really, really > likely." > > Of course the probability that two people of the same age will die in > the same month is higher than the probability that two randomly selected > people will die in the same month. Likelihood of death is not constant > throughout life. Frankly I take the actuaries that suggest my insurance > company up my life insurance premiums every 5 years waaaaaay more > seriously than you. > > If you assume away all the interesting bits, your conclusions get pretty > useless. > > Actually, people study this stuff. See, for example: > > http://biomedgerontology.oxfordjournals.org/content/53A/6/M441.full.pdf I give up. My mathematics professor would call it "existence theorem": Whatever way you will find to simplify your point for digest by one person, there exists another person that will ridicule that sort of presentation. Going back to where it started. This still doesn't convince me that hard drives in my server room will start (or do already) die in pairs. Nope. Doesn't happen. And there are many of them made by same manufacturer, same model, same production date, same batch. Valeri > > --Jon Radel > jon@radel.com > > ++++++++++++++++++++++++++++++++++++++++ Valeri Galtsev Sr System Administrator Department of Astronomy and Astrophysics Kavli Institute for Cosmological Physics University of Chicago Phone: 773-702-4247 ++++++++++++++++++++++++++++++++++++++++
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?26823.128.135.52.6.1461102313.squirrel>