> it can push upto 200Gbps of traffic only.

The limit depends on NICs and ports available for the traffic generator machine.
In CSIT, most testbeds have just one NIC per tested type,
so even with bi-directional traffic we are limited on NIC despite using two 
ports.

The proof is in one special testbed 3na-spr where we have multiple NICs
of the same type (CX7, 4 NICs but two with only one port cabled).
For the usual "one port per direction" tests, using ports on different NIC,
we can achieve [1] ~400 Gbps there with large enough packets,
and it also allows us to run "three ports per direction" tests
achieving [2] ~800 Gbps.
The topology file is here [3] and direct link to telemetry
of randomly selected tests are [4] and [5].

Those are coverage tests executed only one per release,
as results from single-physical-core tests are more stable for regression 
detection purposes.
But let me know if you are interested in some specific one-off results I could 
produce.

Vratko.

[1] 
https://csit.fd.io/coverage/#eNpVj8EOgjAMhp9mXszMHLKdOIi8h2mgkSUTmnYivL0sHqaXJv_392tSwYh9CvPUKN8q6znKPlV1PX6DdcZlYG8ZDK9U2oXor1mQS2ndyTjN-3UQ1AJEY2BkoDBIsWjcilJNoIVYW2MeaOncr35BSPoZ17oowAjFCXTJwXcHGef3PULCqd8aVedfflHeqrsPJbJELA
[2] 
https://csit.fd.io/coverage/#eNpVj8EKwjAMhp-mXqRSOteedlD3HhK24Ap1C0mt29u74qF6Cfzfny8QwYhDCsvcKX9V1nOUfarmcvwG64wrwN4KGF-ptpnor8nItbTuZJzm_ToIagGiKTAyUBilWjRtVWlm0EKsrTEPdNQMq88IST_j2lYFGKE6gc4l-P4g0_K-R0g4D1un2vLLLypbbf8BJ7JEMg
[3] 
https://github.com/FDio/csit/blob/rls2606/topologies/available/lf_3na_spr_testbed21.yaml
[4] 
https://logs.fd.io/vex-yul-rot-jenkins-1/csit-vpp-perf-report-coverage-2606-3na-spr/28635430896/log.html.gz#s1-s1-s1-s1-s2-t5-k2-k15-k20-k14-k1-k1-k1-k1
[5] 
https://logs.fd.io/vex-yul-rot-jenkins-1/csit-vpp-perf-report-coverage-2606-3na-spr/28566962521/log.html.gz#s1-s1-s1-s1-s8-t4-k2-k10-k20-k14-k1-k1-k1-k1

-----Original Message-----
From: [email protected] <[email protected]> On Behalf Of Dinesh via 
lists.fd.io
Sent: Monday, 31 August, 2026 13:13
To: [email protected]
Subject: Re: [vpp-dev] Query on 400-gig throughput

Thanks all for your reply.

I have seen in Trex v3.0, that it can push upto 200Gbps of traffic only. 
So will it support 400Gbps of traffic using Trex and we need to optimize 
anything in startup.conf ?

Thanks

On 31/08/26 15:22, Zhang, Fan via lists.fd.io wrote:
> The latest 1.8Tbps Intel IPsec whitepaper detailed the setup guide and 
> IXIA packet generator configurations.
>
> https://builders.intel.com/solutionslibrary/intel-avx-512-high-performance-ipsec-with-4th-gen-intel-xeon-scalable-processor-technology-guide
>  
>
>
>
> Regards,
>
> Fan
>
> On 31/08/2026 09:18, Benoit Ganne (bganne) via lists.fd.io wrote:
>>> Just wanted to check whether vpp supports 400gig of throughput ?
>> Yes, 1+Tbps of IPsec was demonstrated a few years back for example, 
>> search for Intel whitepapers on VPP.
>>
>>> If yes, which will be the better traffic generator to test 400gig 
>>> traffic.
>> Trex should be able to generate that with a big enough machine. Other 
>> options should work as well.
>>
>> best
>> ben
>>
>>
>>
>>
>
> 
>
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#27157): https://lists.fd.io/g/vpp-dev/message/27157
Mute This Topic: https://lists.fd.io/mt/121010181/21656
Group Owner: [email protected]
Unsubscribe: https://lists.fd.io/g/vpp-dev/leave/14379924/21656/631435203/xyzzy 
[[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-

  • ... Dinesh via lists.fd.io
    • ... Benoit Ganne (bganne) via lists.fd.io
      • ... Zhang, Fan via lists.fd.io
        • ... Dinesh via lists.fd.io
          • ... Vratko Polak -X (vrpolak - PANTHEON TECHNOLOGIES at Cisco) via lists.fd.io
            • ... Dinesh via lists.fd.io
              • ... Vratko Polak -X (vrpolak - PANTHEON TECHNOLOGIES at Cisco) via lists.fd.io
    • ... Pim van Pelt via lists.fd.io

Reply via email to