Re: [lmi] Retooling: timing comparison

[Vadim Zeitlin]

On Mon, 8 Feb 2016 17:37:46 +0000 Greg Chicares <address@hidden> wrote:

GC> On 02/08/2016 02:49 AM, Vadim Zeitlin wrote:
GC> > On Mon, 8 Feb 2016 02:19:33 +0000 Greg Chicares <address@hidden> wrote:
GC> [...]
GC> > GC> log-20160131T1406Z-gwc 20160131T1500Z  54  68% --jobs=2'
GC> > GC> log-20160130T2335Z-OLD 20160131T0039Z  64  81% --jobs=2
GC> > ...
GC> > GC> The "gwc" machine is 2 x E5-2630 v3 with a 1TB Samsung 850 pro SSD.
GC> > GC> The "OLD" (2009) machine is 2 x E5520 with a 2TB WDC WD2003FZEX HDD.
GC> > GC> Both these machines were running msw-xp in a kvm-qemu VM, hosted by
GC> > GC> debian-7.
GC> > 
GC> >  I just don't understand these results at all :-( Leaving aside the office
GC> > machines which have much slower disks, the fact that you gain only 10
GC> > minutes or ~15% when switching from "OLD" to "gwc" with -j2 is just
GC> > incomprehensible. Whether the bottleneck is IO or CPU, you should have
GC> > gained more than this.
GC> 
GC> Better hardware, but the same 32-bit msw-xp VM. The gain is much
GC> greater for cross-compiling with no VM...

 OK, so the bottleneck is either the VM itself, i.e. XP, or VM software. As
much as I like criticizing Microsoft, I'm afraid it must be the latter...

GC> That's the time it takes to run 'install_msw.sh' (adjusted for cross
GC> compiling),

 Do you plan to commit this modified version (perhaps under a different
name?)? I'd like to test using it too.

GC>   *** comparison ***
GC>     --jobs=8 : 8:50 / 5:40 = 530s / 340s = 160% speed
GC>     --jobs=16: 7:15 / 4:00 = 435s / 240s = 180% speed
GC> 
GC> So the new machine's about 70% faster than the old. Is that
GC> still less than you were expecting?

 No, I think this is pretty reasonable, in absolute terms you go from the
best time of 7:13 to 3:30, and twice as fast is good enough and well worth
the upgrade IMO.

 Regards,
VZ