On Thu, Oct 18, 2012 at 10:36 PM, Robert Haas <robertmhaas@gmail.com> wrote:
> Sadly, the situation on Windows doesn't look so good. I
> don't remember the exact numbers but I think it was something like 40
> or 60 or 80 times slower on the Windows box one of my colleagues
> tested than it is on Linux.
Do you happen to know the hardware and Windows version? Windows
QueryPerformanceCounter that instr_time.h uses should use RDTSC based
timing when the hardware can support it, just like Linux. I don't know
if Windows can avoid syscall overhead though.
> Maybe it's worth finding a platform where
> pg_test_timing reports that timing is very slow and then measuring how
> much impact this has on something like a pgbench or pgbench -S
> workload.
This can easily be tested on Linux by changing to the hpet or acpi_pm
clocksource. There probably still are platforms that can do worse than
this, but probably not by orders of magnitude.
Regards,
Ants Aasma
--
Cybertec Schönig & Schönig GmbH
Gröhrmühlgasse 26
A-2700 Wiener Neustadt
Web: http://www.postgresql-support.de