From mboxrd@z Thu Jan 1 00:00:00 1970 From: Oleksandr Natalenko Subject: Re: TCP and BBR: reproducibly low cwnd and bandwidth Date: Fri, 16 Feb 2018 18:00:21 +0100 Message-ID: <3239348.dk1SAWKVVl@natalenko.name> References: <1697118.nv5eASg0nx@natalenko.name> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Cc: Eric Dumazet , Eric Dumazet , "David S. Miller" , Netdev , Yuchung Cheng , Soheil Hassas Yeganeh , Jerry Chu To: Neal Cardwell Return-path: Received: from vulcan.natalenko.name ([104.207.131.136]:42164 "EHLO vulcan.natalenko.name" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755771AbeBPRAY (ORCPT ); Fri, 16 Feb 2018 12:00:24 -0500 In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: Hi! On p=C3=A1tek 16. =C3=BAnora 2018 17:45:56 CET Neal Cardwell wrote: > Eric raises a good question: bare metal vs VMs. >=20 > Oleksandr, your first email mentioned KVM VMs and virtio NICs. Your > second e-mail did not seem to mention if those results were for bare > metal or a VM scenario: can you please clarify the details on your > second set of tests? Ugh, so many letters simultaneously=E2=80=A6 I'll answer them one by one if= you don't=20 mind :). Both the first and the second set of tests were performed on 2 KVM VMs, but= =20 from now I'll test everything using real HW only to exclude potential=20 influence of virtualisation. Also, as I've already pointed out, on the real= HW=20 the difference is even bigger (~10 times). Now, I'm going to answer other emails of yours, including the actual result= s=20 from the real HW and tcpdump output as requested. Thanks! Regards, Oleksandr