On 6/30/23 09:25, Frode Nordahl wrote: > Hello all, > > On Tue, May 30, 2023 at 5:16 PM Felix Huettner > <felix.huettner@mail.schwarz> wrote: >> >> Hi Dumitru, >> >> On Fri, May 26, 2023 at 01:30:54PM +0200, Dumitru Ceara wrote: >>> On 5/24/23 09:37, Felix Huettner wrote: >>>> Hi everyone, >>> >>> Hi Felix, >>> >>>> >>>> Ilya mentioned to me that you will want to bring openstack examples to >>>> ovn-heater. >>>> >>> >>> Yes, we're discussing that. >>> >>>> I wanted to ask how to best join this effort. It would be great for us >>> >>> Everyone is welcome to contribute! :) >>> >> >> Great, we will try that out. > > Thank you for your interest in this effort, as promised I'm in the > process of organizing a community meeting to get this going. > > Below are some proposals for dates and times, please respond with a > prioritized list of what works best for you and we'll try to land on a > single date/time for a first meeting: > Wednesday July 5th 13:00 UTC > Tuesday July 11th 13:30 UTC > Wednesday July 12th 8:30 UTC >
Wednesday July 12th 8:30 UTC Wednesday July 5th 13:00 UTC Tuesday July 11th 13:30 UTC (I only have a 30 min slot available here) Thanks for organizing this, Frode! >>>> to have a tool to scale test ovn. Also we have an ovn deployment with >>>> currently 450 compute nodes where we can try to extract the typically >>>> used northbound entries from. >>>> >>> >>> This is great! I'm not an OpenStack user so I'm probably not using the >>> right terminology and I might be missing things but I think we're >>> interested in: >>> - how many tenant networks do you have in the cluster >>> - for each tenant network: >>> - how many VMs (on average but maybe also min/max/etc) >>> - how many floating IPs (dnat-and-snat) >>> - how many compute nodes run VMs from this network >>> - for each compute: >>> - how many VMs run on it (on average but also min/max/etc) >>> >> >> Ok, so i'll try to give an overview below. >> If you need any more detail just ask. >> >> General amount of nodes: >> * ~500 Hypervisors >> * 9 Nodes that are just serve as gateway chassis (connected to 4 >> different external networks) >> >> Amount of tenant Networks (== logical routers): 2300 (only 1400 are >> actually hosting VMs). >> >> Amount of logical Routers: 2010 (2000 of these have a chassisredirect >> port that is hosted on one of the 9 nodes. >> >> * Amount of VMs: >> * min: 1 >> * max: 300 >> * avg: 4.5 >> * 630 networks only have a single VM, 230 have two VMs and 190 have 3 >> VMs. >> * Amount of Hypervisors per network: >> * min: 1 >> * max: 90 >> * avg: 3.6 >> * 650 networks only strech on a single hypervisor, 250 on two >> hypervisors, 180 on three hypervisors >> >> Regarding floating ips we have currently 1530. I'll give you the amount >> per Logical router: >> * min: 1 >> * max: 230 >> * avg: 2.7 >> * 430 routers have a single floating ip, 140 routers have 2 floating >> ips. >> >> Regarding VMs per Hypervisor we have: >> * min: 1 >> * max: 80 >> * avg: 15 >> >> >> Other potentially interesting values: >> >> * Amount of Load_Balancer: 0 >> * Amount of Logical_Switch_Port type=virtual: 800 >> * Amount of Port_Group: 6000 >> * Amount of ACL: 35500 >> * All of these referce portgroups in `match` using `@` or `$` >> * Referencing other portgroups using `$`: 8300 > > Thanks alot for providing this information, and gauging the size of > your deployment, your inputs would be very valuable in this effort. > One of the main goals of ovn-heater is the ability to simulate > workloads, because as developers, we seldom have access to actual 500- > or 1000- node clusters. Having a participant in the project with > actual hardware installation of such a size would give us the > opportunity to help confirm the accuracy of our simulations. > +1 _______________________________________________ dev mailing list d...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-dev
