[clue-talk] BAARF - Battle Against Any Raid Five (er, Four, er Free.)

[Jed S. Baer]

On Fri, 19 Oct 2007 19:44:28 -0600
Nate Duehr wrote:

> Good technical detail about why RAID-5 isn't always (in fact rarely)
> the correct technical solution for disk redundancy...
> 
> http://www.miracleas.com/BAARF/BAARF2.html

I remember reading Cary Millsap's articles way back when. But I have to
wonder, if RAID5 is so bad, why is it still so popular? If you have lots
of cache, and things like battery-backed cache as in EMC frames, and
managed storage arrays (such as from HP) which do their own health checks
and can be queried via SNMP and/or configured to send snmp traps to a
monitoring system (see all this stuff I'm finding out about up there at
HP?), is it really as bad as these guys say? Have some of the objections
about SCSI drives and sector relocations been addressed, either through
SMART capable drives (there's that health check stuff again) or in the
movement to SAS?

I configure storage all the time. Not with any particular intent, other
than to have LUNs available for test purposes, but it's almost a daily
thing. Whether it's SmartArrays or FibreChannel, I haven't seen RAIDs 2-4
anywhere.

jed