Re: Help clear up a misconception about SMP systems

Alan Cox (alan@lxorguk.ukuu.org.uk)
Sat, 10 Oct 1998 00:07:45 +0100 (BST)

> When two or more processors come into play, what is the increase rate of
> processing over a single processor? Is does it works out as 100% increase
> over that of a single processor or is it limited by other factors such as
> caching, disk i/o or whatever?
>
> I'm guessing it's a 100% increase.

With the perfect application match its 99.9% or so. In the real world
the rule people normally quote for machines of that type is to expect a
"good" result of 100%+80%+60%+40% ... by CPU.

Linux 2.0 is a bit worse than that quite often 2.1.x a bit better.

For most PC's the problem is memory bandwidth. Most single CPU x86 PC's are
horribly memory bandwidth starved - not suprisingly since memory bandwidth
is very expensive compared to clock speed and other enhancements end users
understand.

Cache bound stuff is materially better SMP than stuff using main memory. In
fact on many machines main memory intensive stuff is the same speed single
or dual CPU

Alan

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/