On Tue, Dec 20, 2011 at 01:50:53PM +0100, Dmitrij D. Czarkoff wrote: > On Tue, Dec 20, 2011 at 10:07:22AM +0100, Francois Tigeot wrote: > > > > Since VFS accounting uses a per mount-point spinlock to protect its > > counters, > > there is some concern it will degrade write performance on SMP machines. > > > > To measure if this was the case or not, I've run some blogbench benchmarks > > on > > a small Xeon system; the details of the test and the results are in the > > attached PDF. > > I've made a plot which shows the results in more evident way (if I correctly > understand the test results data).
Heh, nice way to represent the variations, thanks :) I've done a few more tests with blogbench and I sometimes get more than 12% variations with the exact same software and hardware. I'm now convinced blogbench was not a good tool and I'm currently looking for another benchmark. -- Francois Tigeot
