FreeBSD and "make -j# buildworld" usability

[Kent Stewart]

On Monday 16 October 2006 12:46, Nikolas Britton wrote:
> On 10/13/06, Kent Stewart <kstewart at owt.com> wrote:
> > On Friday 13 October 2006 07:31, Buki wrote:
> > > Hi,
> > >
> > > I searched the archives and web a little but found many different
> > > opinions on stability/usability of using make -j# with buildworld
> > > (and buildkernel).
> > >
> > > So I am asking if it is a good idea to use make -j on production
> > > boxes.
> >
> > I tested buildworlds with different values for -j. On single
> > processors, using a script that basically looked like
> >
> > time make -j? ...
> >
> > yielded fastest builds when I didn't specify a value for -j. On
> > dual cpu's a value around -j8 yielded the fastest build.
>
> That's odd, your results don't jive with this:
> http://people.freebsd.org/~fsmp/SMP/akgraph-a/graph1.html

His results are too old to use as a reference. My system was the next 
step up and is also too old to consider current. His cpus were Intel 
Pentium pro 200MHZ and the bus speed alone (66MHz) would make a big 
difference. Only having 64MB of memory may also skew the results.

My system was 2 Intel pIII 866's and had 256-512MB of SDRAM or DDR 
memory.

Kent 


>
> Although that report is quite old... My general rule of thumb for -j
> is n +1, where n equals the total number of cpu cores. This is
> generally enough to keep to processor(s) occupied without over
> stressing the system. Maybe n * 2 is more appropriate, can you post
> the results from your test?

-- 
Kent Stewart
Richland, WA

http://www.soyandina.com/ "I am Andean project".
http://users.owt.com/kstewart/index.html