Re: [Hampshire] Gentoo adaptec RAD controller hangs

Keith Edmunds <kae@???> wrote:
> RAID-5 is computationally expensive, and as it
> relies on at least three drives as opposed to
> RAID-1's two, it has a shorted MTBF.

Really? Do you have a reference for that?

I can see that RAID 5 is more vulnerable after a
drive failure and before rebuild completes - in
that state, two-drive RAID 1's MTBF will be that
of the single working drive, whereas RAID 5's MTBF
will be combination of the two or more working
drives. That's assuming the rest of the system is
identical.

When operating in redundant mode, don't the two
levels have the same MTBF - again, assuming
otherwise identical systems?

> If you really want RAID-5, use a proper,
> dedicated (read: ???) hardware controller.

I've found Linux's software RAID 5 gives
acceptable performance for small-office servers,
using P3-era hardware.

Given that hardware RAID controllers usually cost
extra, I wonder what the balance is like between
hardware RAID vs. software RAID and spending the
money on faster processors or more memory...

Nick.

--
Nick Chalk ................. once a Radio Designer
Confidence is failing to understand the problem.