Re: [vpp-dev] VPP Performance data
Thanks Dave, Jerome. Will go through it. I was specifically looking on VPP 20.01 as I was able to get FRR 3.0.3 working for IPv4 BGP, OSPF with VPP 20.01. Need to try out FRR BGP neighbors and routes scale with VPP DP, so trying to get VPP performance number for 20.01. But master/latest data would be good. Regards, Kausik From: Dave Barach (dbarach) Sent: Monday, March 30, 2020 12:35 PM To: Jerome Tollet (jtollet) ; Majumdar, Kausik ; vpp-dev@lists.fd.io Subject: RE: [vpp-dev] VPP Performance data Message received from external source. Exercise caution when opening attachments, clicking links, or exchanging information. If you care about master/latest, also check out https://docs.fd.io/csit/master/trending/<https://secure-web.cisco.com/12Y83d56KQbs4CrWWwH01anzw0y8Yk2uFxJNvWIpDHdjbzzLsDeZgdUTA4ELscz6YODQKL_088uBipatYw0vteZB8EPCj4BENIfZqB8eM-bPleQNAH4EDa-UAPJ0D7BdLHNFo8fnP5AC8RBbk2XzCia5vnzD7cNYNoZeq02BpwVxoXL6ZK_A25xDOPLvh9tH3reSA6n7nbLmzVqhncln2r3R5lSTXtroh05RcCydNgKUjEpeDeEES8XUolPWWWEBU345vUEja0tHdoDWw7vOesAty_yWXz5Ra96WbpgHCOxTB-uyh7nWKSr-EbLIQ-b0TM0wAcryVGW8O_n4WKE_vqA/https%3A%2F%2Fdocs.fd.io%2Fcsit%2Fmaster%2Ftrending%2F> From: vpp-dev@lists.fd.io<mailto:vpp-dev@lists.fd.io> mailto:vpp-dev@lists.fd.io>> On Behalf Of Jerome Tollet via Lists.Fd.Io Sent: Monday, March 30, 2020 2:40 PM To: Majumdar, Kausik mailto:kausik.majum...@commscope.com>>; vpp-dev@lists.fd.io<mailto:vpp-dev@lists.fd.io> Cc: vpp-dev@lists.fd.io<mailto:vpp-dev@lists.fd.io> Subject: Re: [vpp-dev] VPP Performance data Hello, You should have a look at https://docs.fd.io/csit/rls2001/report/<https://secure-web.cisco.com/1wwd4id_wYGh_3JojpB_aQfJIZTbADNVcty0AUUZDquDnmDK6dP808y2tLchPQD0cYXo9Dh7XClM3LNXKNrJIdSyhcDKp25lUCWCdJqzisf21Ejt3_AB8Ud_jKiUPgTm56REbNYKCXt7ZAfYpbVKiwXjlj1yDcV7nydnDdbwVFycv0-pnlkSjRvMjy-CUSrAM10FhDbbxRduouu2Bz32ccWWxcGisa7Nt-X6E0o_YhmQWnpNjSKD8W8gkmPCxb40W2rPLnef5ymBTqJa0XgnT9udnscJ6z9Yy77MntxTP9Vbfkuc-Q_ciCS8Y8aAzIX_KHv57nDuOwKdgSpe3Z1AZiA/https%3A%2F%2Fdocs.fd.io%2Fcsit%2Frls2001%2Freport%2F> Jerome De : mailto:vpp-dev@lists.fd.io>> au nom de "Majumdar, Kausik" mailto:kausik.majum...@commscope.com>> Date : lundi 30 mars 2020 à 19:47 À : "vpp-dev@lists.fd.io<mailto:vpp-dev@lists.fd.io>" mailto:vpp-dev@lists.fd.io>> Objet : [vpp-dev] VPP Performance data Hi, Can someone please share the VPP performance measurement data in different packet sizes for plain vanilla IPv4, IPv6, and tunnel encap/decap for IPSec, VxLAN cases. Also do we have VPP packet forwarding data with at least one Service VNF in the same host or in a remote host, where traffic is service chained to run the services. I am considering the below topologies to get some performance data - 1. Tgen --> IPv4/IPv6 --> Host1 (VPP1) <Tunnel (IPSec or VxLAN)-> Host2 (VPP2) --> IPv4/IPv6 --> Tgen 2. Tgen --> IPv4/IPv6 --> Host1 (VPP1) <Tunnel (IPSec or VxLAN)-> Host2 (VPP2) --> Host2 VNF (Service VM) --> Host2 (VPP2) --> IPv4/IPv6 --> Tgen 3. Tgen --> IPv4/IPv6 --> Host1 (VPP1) <Tunnel (IPSec or VxLAN)-> Host2 (VPP2) --> Host3 VNF (Service VM) --> Host2 (VPP2) --> IPv4/IPv6 --> Tgen I am assuming some analysis already performed in the above scenarios with the Number of CPUs, Cores, SR-IOV for VNF forwarding, or OVS in Kernel for bridging to VNF. If we have some pointer of these data for VPP 20.01 release that would be great. Thanks, Kausik -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#15946): https://lists.fd.io/g/vpp-dev/message/15946 Mute This Topic: https://lists.fd.io/mt/72658703/21656 Group Owner: vpp-dev+ow...@lists.fd.io Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
Re: [vpp-dev] VPP Performance data
If you care about master/latest, also check out https://docs.fd.io/csit/master/trending/ From: vpp-dev@lists.fd.io On Behalf Of Jerome Tollet via Lists.Fd.Io Sent: Monday, March 30, 2020 2:40 PM To: Majumdar, Kausik ; vpp-dev@lists.fd.io Cc: vpp-dev@lists.fd.io Subject: Re: [vpp-dev] VPP Performance data Hello, You should have a look at https://docs.fd.io/csit/rls2001/report/ Jerome De : mailto:vpp-dev@lists.fd.io>> au nom de "Majumdar, Kausik" mailto:kausik.majum...@commscope.com>> Date : lundi 30 mars 2020 à 19:47 À : "vpp-dev@lists.fd.io<mailto:vpp-dev@lists.fd.io>" mailto:vpp-dev@lists.fd.io>> Objet : [vpp-dev] VPP Performance data Hi, Can someone please share the VPP performance measurement data in different packet sizes for plain vanilla IPv4, IPv6, and tunnel encap/decap for IPSec, VxLAN cases. Also do we have VPP packet forwarding data with at least one Service VNF in the same host or in a remote host, where traffic is service chained to run the services. I am considering the below topologies to get some performance data - 1. Tgen --> IPv4/IPv6 --> Host1 (VPP1) <Tunnel (IPSec or VxLAN)-> Host2 (VPP2) --> IPv4/IPv6 --> Tgen 2. Tgen --> IPv4/IPv6 --> Host1 (VPP1) <Tunnel (IPSec or VxLAN)-> Host2 (VPP2) --> Host2 VNF (Service VM) --> Host2 (VPP2) --> IPv4/IPv6 --> Tgen 3. Tgen --> IPv4/IPv6 --> Host1 (VPP1) <Tunnel (IPSec or VxLAN)-> Host2 (VPP2) --> Host3 VNF (Service VM) --> Host2 (VPP2) --> IPv4/IPv6 --> Tgen I am assuming some analysis already performed in the above scenarios with the Number of CPUs, Cores, SR-IOV for VNF forwarding, or OVS in Kernel for bridging to VNF. If we have some pointer of these data for VPP 20.01 release that would be great. Thanks, Kausik -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#15943): https://lists.fd.io/g/vpp-dev/message/15943 Mute This Topic: https://lists.fd.io/mt/72658703/21656 Group Owner: vpp-dev+ow...@lists.fd.io Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
Re: [vpp-dev] VPP Performance data
Hello, You should have a look at https://docs.fd.io/csit/rls2001/report/ Jerome De : au nom de "Majumdar, Kausik" Date : lundi 30 mars 2020 à 19:47 À : "vpp-dev@lists.fd.io" Objet : [vpp-dev] VPP Performance data Hi, Can someone please share the VPP performance measurement data in different packet sizes for plain vanilla IPv4, IPv6, and tunnel encap/decap for IPSec, VxLAN cases. Also do we have VPP packet forwarding data with at least one Service VNF in the same host or in a remote host, where traffic is service chained to run the services. I am considering the below topologies to get some performance data - 1. Tgen --> IPv4/IPv6 --> Host1 (VPP1)Host2 (VPP2) --> IPv4/IPv6 --> Tgen 2. Tgen --> IPv4/IPv6 --> Host1 (VPP1) Host2 (VPP2) --> Host2 VNF (Service VM) --> Host2 (VPP2) --> IPv4/IPv6 --> Tgen 3. Tgen --> IPv4/IPv6 --> Host1 (VPP1) Host2 (VPP2) --> Host3 VNF (Service VM) --> Host2 (VPP2) --> IPv4/IPv6 --> Tgen I am assuming some analysis already performed in the above scenarios with the Number of CPUs, Cores, SR-IOV for VNF forwarding, or OVS in Kernel for bridging to VNF. If we have some pointer of these data for VPP 20.01 release that would be great. Thanks, Kausik -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#15942): https://lists.fd.io/g/vpp-dev/message/15942 Mute This Topic: https://lists.fd.io/mt/72658703/21656 Group Owner: vpp-dev+ow...@lists.fd.io Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
