SuperMicro 5013

[Les Mikesell]

On Mon, 2005-08-22 at 08:45, Jonathan Horne wrote:
> I never understood this argument of hardware raid and software raid.
> 
> To me, software raid is operating system dependant.  Linuxraid partitions,
> or Microsoft partitions mirrored across disks.  Load up fdisk, and what do
> you see, but 2 or more independent disks.

To me, software raid takes main CPU cycles to execute.  And the OS
versions generally provide the advantage of being controller
independent.  Pull one of your mirrored drives out, put it in another
machine with a different vendor's controller and come up working.

> However, an hp smart array, the logical drive is configured in the bios of
> the card.  Why would you call other vendors cards "fake raid" when theirs is
> almost the same process?  Just because its not SCSI?  Im not following you
> here.

The HP and other 'real' raid controllers have on-board processing and
the main CPU/driver only does hardware port access to control it and
DMA to transfer data.  The main system doesn't know/care about the
individual drives bundled into the logical volumes.  But, you can't
pull a drive and have it work right in some other controller.

> If fdisk shows 1 logical drive drive to work with, and not independent
> disks, then why is it fake raid?  Was a software driver loaded to make this
> possible?  Is it less reliable?

'Fake' raid hides part of the driver code in bios but still makes the
system CPU do all the work.  This gives you the worst of both worlds
because the OS must have a driver that connects to the bios, the CPU
does all the work, and the disk isn't going to work in another machine
with a different controller.   It's one of those things used to
cheapen a machine for people who don't know any better.

-- 
  Les Mikesell
    lesmikesell at gmail.com