Re: [vpp-dev] About nat44 translation with multi-tenancy

Hello, Thanks Filip, I understand that vrf 0 means all vrf. If I want to do nat 44 translation in multitenancy without specifying ports, should I enable nat 44 with the following command? ``` vpp# nat44 enable inside-vrf 0 outside-vrf 0 ``` Also, if possible, could you share some limitations?

Re: [vpp-dev] [csit-dev] FD.io CSIT-2110 Release Report is published

Dear Maciek, I didn't mean to be rude, but you know how it feels when you see a community of professionals for specific software and no one could help. I know that no one is obliged to help me, I was simply expecting more out of it. Regards Ameen On Thu, Nov 11, 2021 at 4:39 PM Maciek

Re: [vpp-dev] [csit-dev] FD.io CSIT-2110 Release Report is published

Dear Ameen, What specific statement in the email do you disagree with? This is not a research paper, it is FD.io CSIT release report capturing benchmarking and functional test data for v21.10 release. Regarding your comment about "VPP and the system doesn't function”, I checked

Re: [vpp-dev] [csit-dev] FD.io CSIT-2110 Release Report is published

Hi Stanislav, You are welcome. Thanks for spotting, and you are right, the af_xdp driver numbers did a nose dive indeed. It’s visible on xeon skylake per your link references, and also on: - intel xeon cascadelake [2], [3], - arm thunderx2 [4]. Comparison tables [5], [6] provide a proof that

Re: [vpp-dev] [csit-dev] FD.io CSIT-2110 Release Report is published

Thanks for the report! Out of curiosity, what happened to AF_XDP? PPS rate was much bigger on r21.06. [0] - https://s3-docs.fd.io/csit/rls2110/report/vpp_performance_tests/packet_throughput_graphs/ip4-2n-skx-x710.html [1] -

Re: [vpp-dev] [csit-dev] FD.io CSIT-2110 Release Report is published

I would totally disagree. It is disappointing when you base your research paper on VPP and the system doesn't function. I have asked here a million times about configuring ds-lite, no one helped. On Thu, Nov 11, 2021 at 3:19 PM Maciek Konstantynowicz (mkonstan) via lists.fd.io wrote: > Big

Re: [vpp-dev] [csit-dev] FD.io CSIT-2110 Release Report is published

Big thanks to FD.io community contributors and supporters that enabled this data-rich publication. Great job everyone! Cheers, Maciek On 10 Nov 2021, at 18:12, Tibor Frank via lists.fd.io mailto:tifrank=cisco@lists.fd.io>> wrote: Hi All,

[vpp-dev] Understanding on CIR EIR CB EB of policer

hello, I have a query on CIR EIR CB EB values. Can I get some proper info on these. How to calculate CB/EB values for a rate of traffic say CIR = 100 KBPS, traffic speed = 80 KBPS, so all must fall under confirm bucket right ? Also CB is only Committed Burst or Committed Burst Size (CBS) as per

Re: [vpp-dev] project call

"Damjan Marion via lists.fd.io" writes: Today on the project call, we were discussing about doing some changes on those calls. Proposal from few of us who were attending is: - we should make it less formal, stop doing readouts and focus on specific topics to discuss, if there is no topics