Chris, I agree that the MTTF numbers these days are a lot different than when = the article was originally published. But I think that the structure of the calculations has remained pretty constant. I think you're right about how to model RAID level 0+1 (1+0)-it's just a degenerate case of RAID level 5, where G=3D2. I hate to become the bottleneck on getting your work in front of = everyone. There are lots of people on this list who can do the job as well or = better than I can. :) I'll look at it too (time permitting at the time when you send it), but I'd recommend for you to show it on the list as soon as = you're done. Cary Millsap Hotsos Enterprises, Ltd. http://www.hotsos.com * Nullius in verba * Upcoming events: - Performance Diagnosis 101: 1/4 Calgary - SQL Optimization 101: 2/7 Dallas - Hotsos Symposium 2005: March 6-10 Dallas - Visit www.hotsos.com for schedule details... -----Original Message----- From: oracle-l-bounce@xxxxxxxxxxxxx = [mailto:oracle-l-bounce@xxxxxxxxxxxxx] On Behalf Of chris@xxxxxxxxxxxxxxxxxxxxx Sent: Wednesday, December 15, 2004 6:58 AM To: cary.millsap@xxxxxxxxxx Cc: oracle-l@xxxxxxxxxxxxx Subject: RE: Storage array advice anyone? Cary, Thanks for that, the paper is very interesting. Your recollection = regarding the double-whammy is correct. The figures used for MTBF, disk size etc. are all now a bit out of date = so what I'll do (hopefully before Christmas) is redo the calculations using = todays figures and give a few more scenarios (including a comparision with RAID = 10, which I believe can be treated as RAID5 with 2 disks in the set, pls = correct me if that is an incorrect assumption). Then I'll send the results to the = list for info. Chris PS Cary would you mind if I send the results to you for review first = before sending to the list? Quoting Cary Millsap <cary.millsap@xxxxxxxxxx>: > The probabilities are already worked out, and they're publicly = available =3D > in > the paper called "RAID: High-Performance, Reliable Secondary Storage" = =3D > (an > ACM Surveys article) by Messrs. Chen, Lee, Gibson, Katz, and = Patterson. > > Not many people bother to put them into Excel, but when I once played = =3D > with > the numbers a bit, I realized pretty quickly that the probability of = an > outage-causing double-whammy is a lot worse than most people think. = The > article mentions that point specifically, if I remember correctly. > > The key idea is that the failures of two disks in an array are =3D > frequently > not independent events. Often, the event that just screwed up disk #1 = =3D > has a > higher probability now of screwing up disk #2 before you can fix #1. > > > Cary Millsap > Hotsos Enterprises, Ltd. > http://www.hotsos.com > * Nullius in verba * > > Upcoming events: > - Performance Diagnosis 101: 1/4 Calgary > - SQL Optimization 101: 2/7 Dallas > - Hotsos Symposium 2005: March 6-10 Dallas > - Visit www.hotsos.com for schedule details... > > > -----Original Message----- > From: oracle-l-bounce@xxxxxxxxxxxxx =3D > [mailto:oracle-l-bounce@xxxxxxxxxxxxx] > On Behalf Of Jared Still > Sent: Tuesday, December 14, 2004 3:58 PM > To: chris@xxxxxxxxxxxxxxxxxxxxx > Cc: Stephen.Lee@xxxxxxxx; oracle-l@xxxxxxxxxxxxx > Subject: Re: Storage array advice anyone? > > Chris Dunscombe Christallize Ltd -- //www.freelists.org/webpage/oracle-l -- //www.freelists.org/webpage/oracle-l