If the system in question is idle, I would agree that a modern single or
dual processor machine probably has plenty of spare cycles to calculate
parity. However, odds are that the machine is doing real work (or you
wouldn't care about RAID 5 and would just use RAID 0) AND has valuable data.
If it's busy doing the "real work" part, there probably isn't a whole lot of
spare cycles around to calculate parity bits. Besides, the StrongARM
processor is allegedly a powerhouse for doing just that type of calculation.
Cheers,
Chris
----- Original Message -----
From: "Brian Pomerantz" <[EMAIL PROTECTED]>
To: "Michael" <[EMAIL PROTECTED]>
Cc: <[EMAIL PROTECTED]>
Sent: Wednesday, March 08, 2000 9:20 PM
Subject: Re: Benchmarking.. how can I get more out of my box?
> On Wed, Mar 08, 2000 at 05:38:29PM -0800, Michael wrote:
> >
> > Consider that the poor little controller chip on the raid card is
> > vastly underpowered for what you are asking it to do in raw IO speed
> > plus handling all the raid calculations. Compare that to the excess
> > number crunching capacity of your primary cpu + mmx processor (used
> > for raid parity calculations) and you might find software raid to be
> > a significant improvement if you can get the controller to just
> > concentrate on delivering raw IO to the dma of your main machine.
> >
>
> The "poor little controller" has a > 200MHz StrongArm in it. Which
> since all it does is calculates parity, it should be plenty for the
> three channels it has on it.
>
>
> BAPper
>
>
