Re: [Ecn-sane] [bbr-dev] duplicating the BBRv2 tests at iccrg in flent?

[Neal Cardwell <ncardwell@google.com>]

[-- Attachment #1: Type: text/plain, Size: 1539 bytes --]

On Fri, Apr 5, 2019 at 11:51 AM Dave Taht <dave.taht@gmail.com> wrote:

>
> Was GRO/GSO enabled on the router? host? server?
>

In this particular invocation of this particular test, the sender,
receiver, and router functionality were all running on the same machine,
using network namespaces and veth devices; TSO and GRO were enabled.


> >
> >>
> >> 5) What was the result with fq_codel instead?
> >
> >
> > With fq_codel and the same ECN marking threshold (fq_codel ce_threshold
> 242us), we see slightly smoother fairness properties (not surprising) but
> with slightly higher latency.
> >
> > The basic summary:
> >
> > retransmits: 0
> > flow throughput: [46.77 .. 51.48]
> > RTT samples at various percentiles:
> >   %   | RTT (ms)
> > ------+---------
> >    0    1.009
> >   50    1.334
> >   60    1.416
> >   70    1.493
> >   80    1.569
> >   90    1.655
> >   95    1.725
> >   99    1.902
> >   99.9  2.328
> >  100    6.414
>
> This is lovely. Is there an open source tool you are using to generate
> this from the packet capture? From wireshark? Or is this from sampling
> the TCP_INFO parameter of netperf?


Thanks. The results and bandwidth graphs are from an internal test
orchestration/evaluation/visualization tool written a few years ago by our
BBR team member, Soheil Hassas Yeganeh, and further enhanced by others on
our team over the years. We are trying to find the time to open-source it,
but haven't yet. It can generate the graphs either from pcap files or "ss"
output. This one was from "ss" output.

neal

[-- Attachment #2: Type: text/html, Size: 2277 bytes --]