<div dir="ltr"><div>Thanks for kicking this off. I've left some comments in the review.<br><br></div>-K<br></div><div class="gmail_extra"><br><div class="gmail_quote">On Tue, Aug 8, 2017 at 10:07 AM, Matthieu Simonin <span dir="ltr"><<a href="mailto:matthieu.simonin@inria.fr" target="_blank">matthieu.simonin@inria.fr</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hello,<br>
<br>
As discussed in the last meeting, I started to formalize the content of the etherpad in the performance WG documentation :<br>
<br>
<a href="https://review.openstack.org/#/c/491818/" rel="noreferrer" target="_blank">https://review.openstack.org/#<wbr>/c/491818/</a><br>
<br>
I've set some co-authorship according to what I saw in the etherpad. I guess this list can be shrunk/expanded on demand :)<br>
<span class=""><br>
Best,<br>
<br>
Matt<br>
<br>
<br>
----- Mail original -----<br>
</span>> De: "Matthieu Simonin" <<a href="mailto:matthieu.simonin@inria.fr">matthieu.simonin@inria.fr</a>><br>
<span class="">> À: "OpenStack Development Mailing List (not for usage questions)" <<a href="mailto:openstack-dev@lists.openstack.org">openstack-dev@lists.<wbr>openstack.org</a>><br>
</span>> Envoyé: Jeudi 6 Juillet 2017 16:31:46<br>
<div class="HOEnZb"><div class="h5">> Objet: Re: [openstack-dev] [FEMDC][MassivelyDistributed] Strawman proposal for message bus analysis<br>
><br>
> ----- Mail original -----<br>
> > De: "Paul-Andre Raymond" <<a href="mailto:paul-andre.raymond@nexius.com">paul-andre.raymond@nexius.com</a><wbr>><br>
> > À: "OpenStack Development Mailing List (not for usage questions)"<br>
> > <<a href="mailto:openstack-dev@lists.openstack.org">openstack-dev@lists.<wbr>openstack.org</a>><br>
> > Envoyé: Mercredi 5 Juillet 2017 21:48:29<br>
> > Objet: Re: [openstack-dev] [FEMDC][MassivelyDistributed] Strawman proposal<br>
> > for message bus analysis<br>
> ><br>
> > Thank you Matt,<br>
> ><br>
> > This is very insightful. It helps.<br>
> ><br>
> > The second link did not work for me.<br>
><br>
> Oh yeah that's probably because of the on-going doc migration[1].<br>
><br>
> [1]:<br>
> <a href="http://specs.openstack.org/openstack/docs-specs/specs/pike/os-manuals-migration.html" rel="noreferrer" target="_blank">http://specs.openstack.org/<wbr>openstack/docs-specs/specs/<wbr>pike/os-manuals-migration.html</a><br>
><br>
> ><br>
> > In the presentation, it mentioned that the load consisted “Boot and List”<br>
> > operations through Rally.<br>
> > Did I understand well?<br>
><br>
> Yes.<br>
><br>
> > Were those hitting the Openstack UI?<br>
><br>
> Rally benchmarks put loads the various APIs and gather some metrics about<br>
> the execution (time, failures...)<br>
><br>
> > Was keystone involved? Was it using Fernet or another sort of token?<br>
><br>
> Keystone is indeed involved and the token was at that time UUID token.<br>
><br>
> ><br>
> > Intuitively, I expected<br>
> > - the big driver for performance on mariadb would be authentication tokens.<br>
> > And fernet would allow to control that.<br>
> > - The big driver for performance on rabbitmq would be ceilometer, and it is<br>
> > not clear from your presentation that any telemetry data hit the message<br>
> > queue.<br>
><br>
> Telemetry wasn't set up, I guess this would have killed Rabbit earlier in the<br>
> tests.<br>
> The split between notification and RPC messaging is interesting in this area.<br>
><br>
> Bye,<br>
><br>
> Matt<br>
><br>
> ><br>
> > Regards,<br>
> ><br>
> > Paul-Andre<br>
> ><br>
> ><br>
> ><br>
> > -----Original Message-----<br>
> > From: Matthieu Simonin <<a href="mailto:matthieu.simonin@inria.fr">matthieu.simonin@inria.fr</a>><br>
> > Reply-To: "OpenStack Development Mailing List (not for usage questions)"<br>
> > <<a href="mailto:openstack-dev@lists.openstack.org">openstack-dev@lists.<wbr>openstack.org</a>><br>
> > Date: Saturday, July 1, 2017 at 4:42 AM<br>
> > To: "OpenStack Development Mailing List (not for usage questions)"<br>
> > <<a href="mailto:openstack-dev@lists.openstack.org">openstack-dev@lists.<wbr>openstack.org</a>><br>
> > Subject: Re: [openstack-dev] [FEMDC][MassivelyDistributed] Strawman<br>
> > proposal<br>
> > for message bus analysis<br>
> ><br>
> > Hi Paul-André,<br>
> ><br>
> > This was without ceilometer. Nova + Neutron were consuming a lot of<br>
> > connections.<br>
> > Some charts are available in the Barcelona presentation[1] and the<br>
> > performance docs[2].<br>
> > In the latter you'll find some telemetry related tests.<br>
> ><br>
> > [1]:<br>
> > <a href="https://www.openstack.org/assets/presentation-media/Chasing-1000-nodes-scale.pdf" rel="noreferrer" target="_blank">https://www.openstack.org/<wbr>assets/presentation-media/<wbr>Chasing-1000-nodes-scale.pdf</a><br>
> > [2]: <a href="https://docs.openstack.org/developer/performance-docs/" rel="noreferrer" target="_blank">https://docs.openstack.org/<wbr>developer/performance-docs/</a><br>
> ><br>
> > Best,<br>
> ><br>
> > Matt<br>
> ><br>
> > ----- Mail original -----<br>
> > > De: "Paul-Andre Raymond" <<a href="mailto:paul-andre.raymond@nexius.com">paul-andre.raymond@nexius.com</a><wbr>><br>
> > > À: "OpenStack Development Mailing List (not for usage questions)"<br>
> > > <<a href="mailto:openstack-dev@lists.openstack.org">openstack-dev@lists.<wbr>openstack.org</a>><br>
> > > Envoyé: Vendredi 30 Juin 2017 18:42:04<br>
> > > Objet: Re: [openstack-dev] [FEMDC][MassivelyDistributed] Strawman<br>
> > > proposal for message bus analysis<br>
> > ><br>
> > > Hi Matthieu,<br>
> > ><br>
> > > You mentioned 15000 connections with 1000 compute nodes.<br>
> > > Was that mostly Nova? Was ceilometer involved?<br>
> > > I would be curious to know how much AMQP traffic is Control<br>
> > > related<br>
> > > (e.g. spinning up VMs) vs how much is telemetry related in a<br>
> > > typical<br>
> > > openstack deployment.<br>
> > > Do we know that?<br>
> > ><br>
> > > I have also left some comments in the doc.<br>
> > ><br>
> > > Paul-Andre<br>
> > ><br>
> > ><br>
> > > -----Original Message-----<br>
> > > From: Matthieu Simonin <<a href="mailto:matthieu.simonin@inria.fr">matthieu.simonin@inria.fr</a>><br>
> > > Reply-To: "OpenStack Development Mailing List (not for usage<br>
> > > questions)"<br>
> > > <<a href="mailto:openstack-dev@lists.openstack.org">openstack-dev@lists.<wbr>openstack.org</a>><br>
> > > Date: Wednesday, June 21, 2017 at 6:54 PM<br>
> > > To: "OpenStack Development Mailing List (not for usage<br>
> > > questions)"<br>
> > > <<a href="mailto:openstack-dev@lists.openstack.org">openstack-dev@lists.<wbr>openstack.org</a>><br>
> > > Subject: Re: [openstack-dev] [FEMDC][MassivelyDistributed]<br>
> > > Strawman<br>
> > > proposal for message bus analysis<br>
> > ><br>
> > > Hi Ken,<br>
> > ><br>
> > > Thanks for starting this !<br>
> > > I've made a first pass on the epad and left some notes and<br>
> > > questions<br>
> > > there.<br>
> > ><br>
> > > Best,<br>
> > ><br>
> > > Matthieu<br>
> > > ----- Mail original -----<br>
> > > > De: "Ken Giusti" <<a href="mailto:kgiusti@gmail.com">kgiusti@gmail.com</a>><br>
> > > > À: "OpenStack Development Mailing List (not for usage<br>
> > > > questions)"<br>
> > > > <<a href="mailto:openstack-dev@lists.openstack.org">openstack-dev@lists.<wbr>openstack.org</a>><br>
> > > > Envoyé: Mercredi 21 Juin 2017 15:23:26<br>
> > > > Objet: [openstack-dev] [FEMDC][MassivelyDistributed]<br>
> > > > Strawman<br>
> > > > proposal for message bus analysis<br>
> > > ><br>
> > > > Hi All,<br>
> > > ><br>
> > > > Andy and I have taken a stab at defining some test<br>
> > > > scenarios<br>
> > > > for<br>
> > > > anal the<br>
> > > > different message bus technologies:<br>
> > > ><br>
> > > > <a href="https://etherpad.openstack.org/p/1BGhFHDIoi" rel="noreferrer" target="_blank">https://etherpad.openstack.<wbr>org/p/1BGhFHDIoi</a><br>
> > > ><br>
> > > > We've started with tests for just the oslo.messaging layer<br>
> > > > to<br>
> > > > analyze<br>
> > > > throughput and latency as the number of message bus clients<br>
> > > > -<br>
> > > > and<br>
> > > > the bus<br>
> > > > itself - scale out.<br>
> > > ><br>
> > > > The next step will be to define messaging oriented test<br>
> > > > scenarios<br>
> > > > for an<br>
> > > > openstack deployment. We've started by enumerating a few<br>
> > > > of<br>
> > > > the<br>
> > > > tools,<br>
> > > > topologies, and fault conditions that need to be covered.<br>
> > > ><br>
> > > > Let's use this epad as a starting point for analyzing<br>
> > > > messaging -<br>
> > > > please<br>
> > > > feel free to contribute, question, and criticize :)<br>
> > > ><br>
> > > > thanks,<br>
> > > ><br>
> > > ><br>
> > > ><br>
> > > > --<br>
> > > > Ken Giusti (<a href="mailto:kgiusti@gmail.com">kgiusti@gmail.com</a>)<br>
> > > ><br>
> > > > ______________________________<wbr>______________________________<wbr>______________<br>
> > > > OpenStack Development Mailing List (not for usage<br>
> > > > questions)<br>
> > > > Unsubscribe:<br>
> > > > <a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" rel="noreferrer" target="_blank">OpenStack-dev-request@lists.<wbr>openstack.org?subject:<wbr>unsubscribe</a><br>
> > > > <a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" rel="noreferrer" target="_blank">http://lists.openstack.org/<wbr>cgi-bin/mailman/listinfo/<wbr>openstack-dev</a><br>
> > > ><br>
> > ><br>
> > > ______________________________<wbr>______________________________<wbr>______________<br>
> > > OpenStack Development Mailing List (not for usage questions)<br>
> > > Unsubscribe:<br>
> > > <a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" rel="noreferrer" target="_blank">OpenStack-dev-request@lists.<wbr>openstack.org?subject:<wbr>unsubscribe</a><br>
> > > <a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" rel="noreferrer" target="_blank">http://lists.openstack.org/<wbr>cgi-bin/mailman/listinfo/<wbr>openstack-dev</a><br>
> > ><br>
> > ><br>
> > ><br>
> > ><br>
> > > ______________________________<wbr>______________________________<wbr>______________<br>
> > > OpenStack Development Mailing List (not for usage questions)<br>
> > > Unsubscribe:<br>
> > > <a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" rel="noreferrer" target="_blank">OpenStack-dev-request@lists.<wbr>openstack.org?subject:<wbr>unsubscribe</a><br>
> > > <a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" rel="noreferrer" target="_blank">http://lists.openstack.org/<wbr>cgi-bin/mailman/listinfo/<wbr>openstack-dev</a><br>
> > ><br>
> ><br>
> > ______________________________<wbr>______________________________<wbr>______________<br>
> > OpenStack Development Mailing List (not for usage questions)<br>
> > Unsubscribe:<br>
> > <a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" rel="noreferrer" target="_blank">OpenStack-dev-request@lists.<wbr>openstack.org?subject:<wbr>unsubscribe</a><br>
> > <a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" rel="noreferrer" target="_blank">http://lists.openstack.org/<wbr>cgi-bin/mailman/listinfo/<wbr>openstack-dev</a><br>
> ><br>
> ><br>
> > ______________________________<wbr>______________________________<wbr>______________<br>
> > OpenStack Development Mailing List (not for usage questions)<br>
> > Unsubscribe: <a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" rel="noreferrer" target="_blank">OpenStack-dev-request@lists.<wbr>openstack.org?subject:<wbr>unsubscribe</a><br>
> > <a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" rel="noreferrer" target="_blank">http://lists.openstack.org/<wbr>cgi-bin/mailman/listinfo/<wbr>openstack-dev</a><br>
> ><br>
><br>
> ______________________________<wbr>______________________________<wbr>______________<br>
> OpenStack Development Mailing List (not for usage questions)<br>
> Unsubscribe: <a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" rel="noreferrer" target="_blank">OpenStack-dev-request@lists.<wbr>openstack.org?subject:<wbr>unsubscribe</a><br>
> <a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" rel="noreferrer" target="_blank">http://lists.openstack.org/<wbr>cgi-bin/mailman/listinfo/<wbr>openstack-dev</a><br>
><br>
<br>
______________________________<wbr>______________________________<wbr>______________<br>
OpenStack Development Mailing List (not for usage questions)<br>
Unsubscribe: <a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" rel="noreferrer" target="_blank">OpenStack-dev-request@lists.<wbr>openstack.org?subject:<wbr>unsubscribe</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" rel="noreferrer" target="_blank">http://lists.openstack.org/<wbr>cgi-bin/mailman/listinfo/<wbr>openstack-dev</a><br>
</div></div></blockquote></div><br><br clear="all"><br>-- <br><div class="gmail_signature" data-smartmail="gmail_signature">Ken Giusti (<a href="mailto:kgiusti@gmail.com" target="_blank">kgiusti@gmail.com</a>)</div>
</div>