On Monday 29 December 2003 05:31 pm, Ged Haywood wrote:
> Hello again,
>
> On Mon, 29 Dec 2003, Kevin Boergens wrote:
> > Sorry, my OT: text was confusing, here is version 2.0
>
> Back on topic now...  :)
>
> > When you do heavy copying between two ATA disks, the CPU load
> > increases sicnificant (or so I think). On a SCSI system, this won't
> > happen.  The card controls the traffic without creating CPU load.
>
> That simply isn't true.  Typically on DMA disc-to-disc copies on my
> IDE drives I'll get *lower* CPU loads than for the SCSI drives on the
> *same* systems.  The copies will of course take longer to complete on
> the IDE drives, I guess the total of CPU cycles burned won't be very
> different but I haven't spent a great deal of time worrying about it.
>
> > What about USB cards?
>
> You can expect the processor load to be very significantly higher
> with USB devices.
>
> 73,
> Ged.

This is interesting as the stable datum has always been that SCSI, with better 
processing abilities and less need of help from the CPU, should run with a 
lower CPU load.

In fact SCSI traffic has a much higher volume of instructions in it than IDE, 
which is processed by the controller and drive. You have f.ex the ability to 
process mutiple requests at the same time, something IDE cannot do.

What kind of number do you have, i.e. what actions results in what CPU load on 
the two?

-- 
Steve

__________________________________________________
You actually need to constantly be alert 
 and willing to handle things, or life 
   will find a way to get you good!


-------------------------------------------------------
This SF.net email is sponsored by: IBM Linux Tutorials.
Become an expert in LINUX or just sharpen your skills.  Sign up for IBM's
Free Linux Tutorials.  Learn everything from the bash shell to sys admin.
Click now! http://ads.osdn.com/?ad_id=1278&alloc_id=3371&op=click
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-users

Reply via email to