[ovirt-devel] [VDSM] rhel6.5 vs rhel7 performance in sampling

[Nir Soffer]

----- Original Message -----
> From: "Francesco Romani" <fromani at redhat.com>
> To: "Nir Soffer" <nsoffer at redhat.com>
> Cc: devel at ovirt.org, "Federico Simoncelli" <fsimonce at redhat.com>
> Sent: Thursday, July 31, 2014 5:28:24 PM
> Subject: Re: [ovirt-devel] [VDSM] rhel6.5 vs rhel7 performance in sampling
> 
> ----- Original Message -----
> > From: "Nir Soffer" <nsoffer at redhat.com>
> > To: "Francesco Romani" <fromani at redhat.com>
> > Cc: devel at ovirt.org, "Federico Simoncelli" <fsimonce at redhat.com>
> > Sent: Thursday, July 31, 2014 4:16:04 PM
> > Subject: Re: [ovirt-devel] [VDSM] rhel6.5 vs rhel7 performance in sampling
> > > Nothing fancy here, just the differences from RHEL6.5 to RHEL7.
> > > See the attached graphs:
> > > * baseline_{cpu,mem}.png is RHEL 6.5
> > > * baseline_{cpu,mem}_7.png is RHEL 7.0
> > > RHEL7 is so *much* better, switching the hypervisor host is by far
> > > the biggest (and easiest) gain we achieved so far.
> > 
> > Are these showing the current vdsm code, before the scalable_sampling
> > branch?
> 
> exactly. Plain vanilla master.
> 
> > > 
> > > coming soon:
> > > - discussion about the impact of the scalable_sampling topic branch
> > 
> > I want to see the numbers the new graphs
> 
> sure, is the very next thing
> 
> > > PS: a couple of updates about specific points (Michal, Nir)
> > > - still no luck with yappi on RHEL7, managed to do the above bench
> > >   with 10 VMs and collected a profile.
> > 
> > Can you share the profile with us?
> 
> please find attached.

Thanks, looking

> 
> > > - 30751 is a win, same of the last changes of
> > > http://gerrit.ovirt.org/#/c/29980/
> > >   (revs 12, 13)
> > 
> > More details?
> 
> Yep, let me just re-run benchmarks to highlight the impact changes on
> isolation.
> 
> After that I'll send a followup mail discussing the impact of
> - our new topic branch
> - our new topic branch + 30751
> on both RHEL 6.5 and 7.0
> 
> not sure if it is worth to bench 29980 v13 directly or to have one set for
> 29980 v11 and one for 29980 v13

I'm sure that you should not waste your time on comparing minor versions
of the patches.

We should focus on comparing master with master + latest scalable_sampling.

Nir