From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-wm0-x236.google.com (mail-wm0-x236.google.com [IPv6:2a00:1450:400c:c09::236]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id 656323BA8E for ; Sun, 5 Nov 2017 14:42:55 -0500 (EST) Received: by mail-wm0-x236.google.com with SMTP id s66so9950610wmf.5 for ; Sun, 05 Nov 2017 11:42:55 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=broadcom.com; s=google; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc; bh=Nt9M7ATcvL4Xu72JTI2+S63wOO7Mp0Rum/rSHphSiIk=; b=ezRDtKZu0m4C8ssetxraLRJz/v2kRrnch5UkA3VW/7u0D8Fmc1Lghn4zV/ZivdbKka OCWkLirSjvUJknaaXJU0FBNOhVIlKXIK2bRG6VnNVfJNAGkxpyudLQLk2d7t9SVVAZpV W0ILZfupAkLHrrW11yh6s4TwREwDGRJt+9MIM= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:cc; bh=Nt9M7ATcvL4Xu72JTI2+S63wOO7Mp0Rum/rSHphSiIk=; b=IzgfRlriOYHsb+67g3TzgsRPk0eengc9CpmXAh2X5YchlYmM26S6hygzV92iN4KIyI 59Qw+D4SHAXkq+mt1ZevwSS8ZZ9i6f8iF8y1O4ujLdBblSUgAK66ZZk0nBjmh7dJi+xP 7bYOhCWjUwLd4hFBu2/xWMwOVxTnUmjyhHSUfIswkZY0pGaz+OVjFdnxNjRRUX9hnBd9 gw8bDOzOjgh9VJkPaXpdx6l3MDB8A8MlXnLELZMWayAzobsZUo7gUbPan83zf+zBxgXz sq71Zf5nBrXZzoKOBm0yL2middvxmuOPNJSiRH0dVmCDZhO4TnKIq/y+/k7A5mu9EV4D DM+w== X-Gm-Message-State: AMCzsaU8CJLZtC70lhaS5DEEz8B9XF5QNrgHYiGpuuzUVA26g8bnkE7E wnnXKU0scNzE43Cc/r9jVVhSqSdEaiK06zH6Nf+0qA== X-Google-Smtp-Source: ABhQp+TAso9QtSggaRqy9P0hjy3621jyfgBrnZiE6OCumSw4b/xJQfsr/jfM/3BOVEqdq8Urp3cSzllk3/CXYQ6pkAE= X-Received: by 10.80.168.129 with SMTP id k1mr17598651edc.109.1509910974019; Sun, 05 Nov 2017 11:42:54 -0800 (PST) MIME-Version: 1.0 Received: by 10.80.186.35 with HTTP; Sun, 5 Nov 2017 11:42:53 -0800 (PST) In-Reply-To: <95AE4A18-0329-42FA-91E2-9EFF356FD8C9@gmail.com> References: <535CD0D7-CB3A-4C42-AB13-548313F9002A@gmail.com> <95AE4A18-0329-42FA-91E2-9EFF356FD8C9@gmail.com> From: Bob McMahon Date: Sun, 5 Nov 2017 11:42:53 -0800 Message-ID: To: Pete Heist Cc: make-wifi-fast@lists.bufferbloat.net Content-Type: multipart/alternative; boundary="94eb2c1959503ff0fb055d418b4f" Subject: Re: [Make-wifi-fast] Flent test hardware X-BeenThere: make-wifi-fast@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 05 Nov 2017 19:42:55 -0000 --94eb2c1959503ff0fb055d418b4f Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable The realtek is definitely not ideal. The test needs to watch the ptp stats to make sure the corrections are stable over the life of the test and throw out bad results per test equipment introducing too much error. Qualifying a NIC for use in test equipment is a bit of a pain. My rationale is to avoid consumer grade products, rather leverage the work of engineers that qualify equipment for data centers, i.e. the data center market is driving the vendor. I find the INTC server class NICs to be the best for this so far. Bob On Sun, Nov 5, 2017 at 5:57 AM, Pete Heist wrote: > > On Nov 5, 2017, at 2:42 AM, Bob McMahon wrote: > > I have some brix with realtek and run ptpd installed with fedora 25. > The corrections are in the 25 microsecond range, though there are > anomalies. These are used for wifi DUTs that go into RF enclosures. > > [root@hera ~]# tail -n 1 /var/log/ptpd2.stats > 2017-11-04 18:33:46.723476, slv, 0cc47afffea87386(unknown)/1, > 0.000000000, -0.000018381, 0.000000000, -0.000018463, 1528.032750001, S, > 0.000000000, 0, -0.000018988, 1403, 1576, 17, -0.000018463, 0.000000000 > > For LAN/WAN traffic, I tend to use the intel quad server adapters in a > supermicro mb desktop with 8 or more real cores. (I think the data cente= r > class machines are worth it.) > > > Thanks for the info. I was wondering how large the PTP error would be wit= h > software timestamps, and I see it=E2=80=99s not bad for most purposes. > > Which Realtek Linux driver does your brix use, and is it stable? The r816= 9 > driver=E2=80=99s BQL support was reverted at some point and it doesn=E2= =80=99t look like > that has changed. > > I trust that the extra cores can help, particularly for tests with high > flow counts, but my project budget won=E2=80=99t allow it, and used hardw= are is too > much to think about at the moment. > > Do you (or anyone) know of any problems with running the Flent client and > server on the same box? In the case of the Proliant Microserver, the > Broadcom 5720 adapter should have separate PCI data paths for each NIC. I > guess the bottleneck will still mainly be the CPU. To get some idea of > what's possible on my current hardware, I tried running rrul_be_nflows > tests with the Flent client and server on the same box, through its local > adapter (with MTU set to 1500) with my current Mac Mini (2.26 GHz Core2 D= uo > P7550). I know that doesn=E2=80=99t predict how it will work over Etherne= t, but > it=E2=80=99s a start. > > https://docs.google.com/spreadsheets/d/1MVxGsreiGKNXhfkMIheNFrH_ > GVllFfiH9RU5ws5l_aY/edit#gid=3D1583696271 > > Although total throughput is pretty good for a low-end CPU, I=E2=80=99m n= ot sure > I=E2=80=99d trust the results above 64/64 flows. 256/256 flows was an epi= c fail, > but I won=E2=80=99t be doing that kind of test. > --94eb2c1959503ff0fb055d418b4f Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable
The realtek is definitely not ideal.=C2=A0 =C2=A0The = test needs to watch the ptp stats to make sure the corrections are stable o= ver the life of the test and throw out bad results per test equipment intro= ducing too much error.

Qualifying a NIC for use in test equipment is= a bit of a pain.=C2=A0 My rationale is to avoid consumer grade products, r= ather leverage the work of engineers that qualify equipment for data center= s, i.e. the data center market is driving the vendor.=C2=A0 =C2=A0I find th= e INTC server class NICs to be the best for this so far.

Bob

On Sun, Nov 5= , 2017 at 5:57 AM, Pete Heist <peteheist@gmail.com> wrote:=

On Nov 5, 2017, at 2:4= 2 AM, Bob McMahon <bob.mcmahon@broadcom.com> wrote:

I have some b= rix with realtek and run ptpd installed with fedora 25.=C2=A0 =C2=A0 The co= rrections are in the 25 microsecond range, though there are anomalies.=C2= =A0 These are used for wifi DUTs that go into RF enclosures.=C2=A0=C2=A0

[root@hera ~]# tail -n 1 /var/log/ptpd2.stats
2017-11-04 18:33:46.723476, slv, 0cc47afffea87386(unknown)/1,=C2=A0 0.000= 000000,=C2=A0-0.000018381,=C2=A0 0.000000000, -0.000018463, 1528.032750001, S, 0.000000000, 0, = -0.000018988, 1403, 1576, 17, -0.000018463,=C2=A0 0.000000000
For LAN/WAN traffic, I tend to use the intel quad server adapters in= a supermicro mb desktop with 8 or more real cores.=C2=A0 (I think the data= center class machines are worth it.)

Thanks for the info. I was wondering how large the PTP error w= ould be with software timestamps, and I see it=E2=80=99s not bad for most p= urposes.

Which Realtek Linux driver does your brix= use, and is it stable? The r8169 driver=E2=80=99s BQL support was reverted= at some point and it doesn=E2=80=99t look like that has changed.

I trust that the extra cores can help, particularly for tes= ts with high flow counts, but my project budget won=E2=80=99t allow it, and= used hardware is too much to think about at the moment.

Do you (or anyone) know of any problems with running the Flent clien= t and server on the same box? In the case of the Proliant Microserver, the = Broadcom 5720 adapter should have separate PCI data paths for each NIC. I g= uess the bottleneck will still mainly be the CPU. To get some idea of what&= #39;s possible on my current hardware, I tried running rrul_be_nflows tests= with the Flent client and server on the same box, through its local adapte= r (with MTU set to 1500) with my current Mac Mini (2.26 GHz Core2 Duo P7550= ). I know that doesn=E2=80=99t predict how it will work over Ethernet, but = it=E2=80=99s a start.


Although total throughput is pretty g= ood for a low-end CPU, I=E2=80=99m not sure I=E2=80=99d trust the results a= bove 64/64 flows. 256/256 flows was an epic fail, but I won=E2=80=99t be do= ing that kind of test.

--94eb2c1959503ff0fb055d418b4f--