Vassili Leonov wrote:
>
> Oliver Schindler wrote:
> >You're wrong here. The default timeslice is 20 ms, but X or an IDE- IRQ
> >can block the system up to 200-300 ms.
> Can you please elaborate on HOW (doing what?) can they block the system
> up to 200-300ms - and if that is really necessary - or just sloppiness.
>
> I could never understand quite clearly, why, even on Linux running
> on PC hardware, if there are several users running gcc at the same
> time things become quite sluggish. On the Sun SPARC with same dhrystone
> performance numbers - everything is crisp - what's the real hardware
> bottleneck on PCs?
A lot depends on the PC but typically:
1) The SPARC will have fast SCSI disks. These work much better then
IDE in a muti task environment.
2) SPARC will have 1MB or more high speed ram cache. PCs are lucky
to have 1/2 this and it will be slower cache chips.
3) The new SPARCs use a crossbar switch rather then a bus for system
interconnection and thus have much better internal bandwidth.
4) IMO the Linux scheduler seems to be tuned for good interactive
responce while Solaris seems to do better at background processing.
5) My dual CPU Ultra SPARC with 512MB RAM and 18GB UWSCSI disks and
21" CRT costs $15,000 while my 400Mhz PII Linux box costs $2,800.
you kind of get what you pay for.
--
--Chris Albertson
[EMAIL PROTECTED] Voice: 626-351-0089 X127
Logicon RDA, Pasadena California Fax: 626-351-0699
--- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR
echo "unsubscribe rtl <Your_email>" | mail [EMAIL PROTECTED]
----
For more information on Real-Time Linux see:
http://www.rtlinux.org/~rtlinux/
