Seems vaguely familiar...
http://portal.acm.org/citation.cfm?id=203282&dl=ACM&coll=portal
I'm just happy it's not Reed-Solomon like StorageTek went with a few years back - imagine doing that without some funky hardware over 16TB of storage. I have a feeling the market generally didn't go a bundle on that approach.
Ironically, VXA (now Exabyte) used this dual-parity technique, along with others including R-S(!), in their VXA packet drive tape-drive technology. At least that's how I think their secret sauce works.
For the history buffs, this is how Garth Gibson saw the world of RAID 9 years ago (with a mention of double-disk-failure RAID codes, in case you're wondering):
http://www.pdl.cmu.edu/RAIDtutorial/Sigarch95.pdf
Though, I'm still waiting for someone to make one of these:
http://www.cs.ucsd.edu/groups/gemini/papers/isca97.ps
*grin*
--On 17 March 2004 09:53 -0800 "Strange, Steve" Steve.Strange@netapp.com wrote:
Chris,
This technical article on the NetApp website has a good technical overview:
http://www.netapp.com/tech_library/3298.html
Cheers,
Steve
-----Original Message----- From: Chris Thompson [mailto:cet1@cus.cam.ac.uk] Sent: Wednesday, March 17, 2004 9:41 AM To: Strange, Steve Cc: toasters@mathworks.com Subject: Re: RAID4 Catch-22
Steve Strange writes: [...]
Data ONTAP 6.5 features RAID-DP, or RAID double parity, which makes
this
problem disappear. The RAID group would then have double redundancy, and could handle a media failure during the reconstruct of a single
disk
without data loss or downtime.
Can anyone elaborate on how this works, in detail?
Chris Thompson Email: cet1@cam.ac.uk