mfs and buildworlds on da SunFire x4600

[Kris Kennaway]

On Thu, May 03, 2007 at 03:29:05AM +0800, Mars G. Miro wrote:

> >You definitely should be measuring some difference - you need to use
> >time(1) though instead of just estimating.
> >
> 
> I did use time, here:
> 
> make -j32 buildworld:
> disk
> 3229.975u 8981.269s 18:55.21 1075.6%    3804+2164k 19702+6594io 2158pf+0w
> 
> malloc mfs ( mdmfs -SM -o async,noatime -s 2G md {src, obj} )
> 3069.783u 9546.250s 18:57.58 1109.0%    3844+2147k 18504+4422io 2628pf+0w
> 
> swap mfs ( mdmfs -S -o async,noatime -s 2G md {src, obj} )
> 3061.682u 9682.776s 19:08.90 1109.2%    3849+2164k 13732+4402io 2053pf+0w
>
> make -j16 buildworld:
> disk
> 3218.722u 7855.885s 17:42.26 1042.5%    3833+2267k 16995+6620io 2445pf+0w
> 
> malloc mfs ( mdmfs -SM -o async,noatime -s 2G md {src, obj} )
> 3187.753u 8199.049s 17:57.63 1056.6%    3839+2238k 6890+4436io 2346pf+0w
> 
> swap mfs ( mdmfs -S -o async,noatime -s 2G md {src, obj} )
> 3200.400u 8213.312s 18:06.88 1050.1%    3860+2243k 12800+4397io 2303pf+0w

Yeah, looks to me like you're not using your CPUs at 100% capacity
(because buildworld doesn't parallelize well to 16 CPUs), so you're
neither I/O bound nor CPU-bound on average.  With a more I/O-bound
test (e.g. maybe just a parallel kernel build) you should see a clear
win from using a md.

Kris