Well, I haven't examined this empirically, but as the quantity of disk
drives in an organization continues increasing, so does the probability
of disk failure, any one of which can mean lost time/money, etc. Drive
reliability is likely not increasing at the same rate that density is,
so the likelihood of lost data is probably increasing. Since LAN speeds
continue to increase, it might start making sense now in clusters of
more than a few machines to make each machine less reliant on its own
disk storage (to the point of not at all other than big swap space) and
use the LAN more. On the LAN put the money into a quality shared
resource - a heavy duty UPS'd, etc. RAID system. Especially if a RAID
system is as easy to build/maintain/use as Alan alludes to (don't know -
never built one).
Billy
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/