Some simple facts:
Seagate ST31000640SS (1TB SATA):
AFR (Average Failure Rate) 0.73%
If you have 200 disks, 1.46 die per year!
ECC 10 Bit: if you have 1024 Errors, 1 will be missed!
Non recoverable Read Errors per Bits read: 1 sector per 10^15 bits ~=128Tbyte
read, and you've hit a read error. (more precise: 116.4 Terabyte)
Read, what CERN has to say:
http://storagemojo.com/2007/09/19/cerns-data-corruption-research/
It's even worse than that mentioned above:
Some RAID-5 Controllers simply didn't work!
CERN now uses (not at all yet, but going there) ZFS just for the obvious
reasons! They produce massive data, and need that data to be correct...
Matthias
You (Orvar Korvar) wrote:
> If you value your data, you should reconsider. But if your data is not
> important, then skip ZFS.
>
> File system data corruption test by researcher:
> http://blogs.zdnet.com/storage/?p=169
>
> ZFS data corruption test by researchers:
> http://www.cs.wisc.edu/wind/Publications/zfs-corruption-fast10.pdf
--
Matthias Pfützner | Tel.: +49 700 PFUETZNER | It is clear that an
Lichtenbergstr.73 | mailto:[email protected] | attempt to master the
D-64289 Darmstadt | AIM: pfuetz, ICQ: 300967487 | absurdities of Unix would
Germany | http://www.pfuetzner.de/matthias/ | exhaust anyone. Jim Davis
_______________________________________________
opensolaris-discuss mailing list
[email protected]
