[erlang-questions] Strange optimization result

[Thomas Lindgren]

--- Anders Nygren <anders.nygren@REDACTED> wrote:

> On 10/22/07, Thomas Lindgren
> <thomasl_erlang@REDACTED> wrote:
> >
> > (I think the ultimate limit in most any language
> would
> > be a time of about 0.33 seconds, the sys time. But
> at
> > this point, we are actually within a factor 10 of
> > that.)
> 
> Huh??
> I thought that sys was the time spent in the kernel.

I was thinking of a variation on Amdahl's Law: even
with infinite speedup of the application, you'd still
be stuck with running stuff in the kernel, which
limits potential speedup. For example, even if all
matching work could be done while waiting for more
data to arrive, the task would still need at least
0.33 seconds. But this is basically just a theoretical
rule of thumb.

> I have tested it but on two cores [using local
ets:update_counter/3] makes no
> noticeable
> difference, even though it sounds like a good idea.

Too bad. Well, good luck with it all.

Best,
Thomas


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com