Re: About Triton DMA IDE vs. SCSI performance
bruce@pixar.com (Bruce Perens) writes:
> I didn't say wide fast. On my system I am getting about 5 MBPS on IDE from
> a Quantum Fireball drive, which is as fast as my narrow fast SCSI drive
> on a narrow fast Adaptec AIC-7850 SCSI on the same motherboard (an IWILL
> P54TS). The speed seems to be limited by the disk media, not by any
> difference between IDE and SCSI.
What have you used for testing? I'm curious to test my own system.
Something more sophisticated than hdparm -t?
> No doubt SCSI works better for multiple devices on a single controller.
> However, for my configuration this is probably moot - the Triton
> provides two DMA capable IDE busses each with its own PCI bridge, and I
> have one disk on each.
So from this it sounds like if you have two IDE drives, you are better
off to put one on each IDE bus. Is that true? I looked for some info
about this, but never found any.
Also, I started using the md (RAID0) driver on a couple of machines,
and it seems to work well. However, you can't test performance with
hdparm since it doesn't understand md devices. Is there a good
alternative?
One one machine I striped 2 identical Quantum fireballs, and things
seem good, but the other machine has drives of different makes/sizes.
The md HOWTO says that you can stripe drives of different makes and
sizes, but it doesn't say when it makes sense (performance wise) to do
this. I did it (just to try it out) with a 1GIG Fireball and a 500MB
Caviar. Could this actually be a win? I know that you have two
mechanisms doing the fetching, but if one drives is slower than the
other, it seems that you might have a problem...
Thanks
--
Rob
Reply to: