> The real performance problem with RAID 5 won't show up until a drive > dies and it starts rebuilding I don't agree with that. RAID5 is very slow for random writes, since it needs to :
"The real problem" is when RAID5 loses a drive and goes from "acceptable" kind of slow, to "someone's fired" kind of slow. Then of course in the middle the rebuild, a bad sector is discovered in some place the filesystem has never visited yet on one of the remaining drives, and all hell breaks loose.
RAID6 is only one extra disk... -- Sent via pgsql-performance mailing list (pgsql-performance@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-performance
