Re: Low performance of BBR compared to cubic

I forgot to subscribe #transport mailing list and missed some replies.

> On Nov 19, 2023, at 3:37 AM, Zhenlei Huang <zlei@FreeBSD.org> wrote:
> 
> Hi,
> 
> While test TCP RACK functions, I tested BBR BTW.
> 
> This is a quick test with iperf3 on bare metal ( old MBP i5 2 cores 4 threads ).

Forgot to mention that test is done with lo0, one thread only.
`iperf3 -s`
and
`iperf3 -c 127.0.0.1`

I would prefer that as a baseline.
Generally it should be good enough, since there's no congestion or loss, and the bottleneck will always be CPU .

> The kernel current/15 with debug options disabled. Following is the performance result:
> 
> freebsd:	37.2 Gbits/sec	1.32 MBytes
> RACK:		27.9 Gbits/sec	1.34 MBytes
> BBR:		2.34 Gbits/sec	223 KBytes
> 
> For freebsd and RACK functions the CC is cubic.
> 
> The last column is Cwnd. BBR's Cwnd looks quite small.
> 
> There's also a report on Telegram Taiwan FreeBSD chat group, but without performance details.
> 
> I believe there is something wrong with BBR. This is not a reasonable good performance compared with
> other tcp congestion control algorithms.
> 
> Or am I missing something ?
> 
> Best regards,
> Zhenlei
>