From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-lb0-x236.google.com (mail-lb0-x236.google.com [IPv6:2a00:1450:4010:c04::236]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (Client CN "smtp.gmail.com", Issuer "Google Internet Authority G2" (verified OK)) by huchra.bufferbloat.net (Postfix) with ESMTPS id A97BC21F304 for ; Mon, 30 Mar 2015 18:30:44 -0700 (PDT) Received: by lbcmq2 with SMTP id mq2so1885946lbc.0 for ; Mon, 30 Mar 2015 18:30:42 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=+8SbedRalkxn+wV3odxR3jN5tByJecGefng0uNLGkc4=; b=PfyF9RqUWsZKFfIt4A+aAHM7icgI3vZ55nveF9pxhq4rOP4z+rSP7wFe77dHaH49YK PIXQF1fgTgJIz6KGLrFO1jYzPe6WJuy7o9zipf5jwkP/GYKyQDnFEYM8fYKVJ2McoKtz RIj6l6YgHMwhP795/tLjCceOa5T5VKBFOrZQ3hioRJREl7IqA9TmYPLp4SrsH18S1LCJ PkVkZrs4eFLh9SlbTeOOjiVMUxW3X5nwRDD5FEGanP907JcoPErpW5xfh/9+U+cgBnBd r5okYQZUqcrpTkA9b/Bd9Qw8trCzj+PE9IIGquAZ0tzLyE8Cl+BwWQW+T3kt6CkDIT6h QSwQ== MIME-Version: 1.0 X-Received: by 10.152.21.99 with SMTP id u3mr28637838lae.105.1427765441869; Mon, 30 Mar 2015 18:30:41 -0700 (PDT) Received: by 10.114.200.73 with HTTP; Mon, 30 Mar 2015 18:30:41 -0700 (PDT) In-Reply-To: References: <20150316203532.05BD21E2@taggart.lackof.org> <123130.1426635142@turing-police.cc.vt.edu> <15A0911A-E3B7-440A-A26B-C5E1489EA98B@viagenie.ca> <1426773234.362612992@apps.rackspace.com> <1426796961.194223197@apps.rackspace.com> <5FD20B4A-A7E8-48C0-89F9-E2EB86DED8A6@gmail.com> <755FC1BE-141C-42FD-B3E3-564488982665@gmail.com> <95B0F8DA-7650-4064-BEE6-F0CDF936D33A@gmail.com> Date: Tue, 31 Mar 2015 03:30:41 +0200 Message-ID: From: Pedro Tumusok To: Jonathan Morton Content-Type: multipart/alternative; boundary=089e0149397afc3e5605128b8bc8 Cc: "dpreed@reed.com" , bloat Subject: Re: [Bloat] Latency Measurements in Speed Test suites (was: DOCSIS 3+ recommendation?) X-BeenThere: bloat@lists.bufferbloat.net X-Mailman-Version: 2.1.13 Precedence: list List-Id: General list for discussing Bufferbloat List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 31 Mar 2015 01:31:13 -0000 --089e0149397afc3e5605128b8bc8 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Justin has signed up for this list, I will just copy the question he sent me directly over at DSLReports and let you chime in with your views. -- I've subscribed to the list. Currently the 'saturation' meter is pinging away at an unrelated server ( dslreports.com) probably it should ping away, and with higher frequency, at one of the servers streaming data in? because then there are more likely to be filled buffers en-route? or are the bloated buffers mainly at the customer end of the experience. I can make the saturation meter display RTT directly, and continue during the test as an preference. I don't really want to have it pinging away during the test because it probably slows the result down. Actually I'll have to check that. Definitely on slow lines it would (like GPRS and 3G). tcptrace on the server side of one stream would immediately reveal average and peak RTT and more. I wonder if that is the goal to be shooting for rather than these more indirect measurements. What is the buffer bloat opinion on the ESNet page? =C2=BBfasterdata.es.net/network-tuning =C2=B7=C2=B7=C2=B7 r-bloat/ they say more not less buffers are needed for 10gig, and its only a problem with residential. On Mon, Mar 30, 2015 at 4:20 PM, Pedro Tumusok wrote: > > > On Mon, Mar 30, 2015 at 4:18 PM, Jonathan Morton > wrote: > >> >> > On 30 Mar, 2015, at 16:56, Pedro Tumusok >> wrote: >> > >> > Also would you mind doing another test and share the link for result, >> so he can check closer? >> >> I still have the result link for the earlier one: >> >> http://www.dslreports.com/speedtest/193440 >> >> Here=E2=80=99s another result without using the webcache, but keeping ev= erything >> else the same: >> >> http://www.dslreports.com/speedtest/199524 >> >> For context, this is through the PowerBook which is running cake, 8Mbps >> down and 1Mbps up. This is somewhat less than the capacity of the link, >> but since it=E2=80=99s 3G it varies quite a lot anyway. >> >> > on the other things, yes I'd like to hear about buffer bloat ideas. >> >> The number-one desirable feature is to carry on measuring the latency >> while the throughput test is ongoing. This is relevant to anyone who wa= nts >> to make a VoIP call or play an online game while using the connection fo= r >> something else. If you have several people in one household, it=E2=80= =99s often >> hard to coordinate their activity to avoid such multitasking. >> >> - Jonathan Morton >> >> > I will forward it to him, hopefully he will signup here also, so the smar= t > people can talk to each other and not having me mess it up :) > > > -- > Best regards / Mvh > Jan Pedro Tumusok > > --=20 Best regards / Mvh Jan Pedro Tumusok --089e0149397afc3e5605128b8bc8 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
Justin has signed up for this list, I will just copy the q= uestion he sent me directly over at DSLReports and let you =C2=A0chime in w= ith your views.

--

I'v= e subscribed to the list.

Currently the 'satur= ation' meter is pinging away at an unrelated server (dslreports.com)
probably it should ping away,= and with higher frequency, at one of the servers streaming data in? becaus= e then there are more likely to be filled buffers en-route?
or ar= e the bloated buffers mainly at the customer end of the experience.

I can make the saturation meter display RTT directly, and= continue during the test as an preference. I don't really want to have= it pinging away during the test because it probably slows the result down.= Actually I'll have to check that. Definitely on slow lines it would (l= ike GPRS and 3G).

tcptrace on the server side of o= ne stream would immediately reveal average and peak RTT and more. I wonder = if that is the goal to be shooting for rather than these more indirect meas= urements.

What is the buffer bloat opinion on the = ESNet page?

=C2=BBfasterdata.es.net/network-tuning =C2=B7=C2=B7=C2=B7= r-bloat/

they say more not less buffers are neede= d for 10gig, and its only a problem with residential.=C2=A0

On Mon, Mar 30,= 2015 at 4:20 PM, Pedro Tumusok <pedro.tumusok@gmail.com> wrote:


On = Mon, Mar 30, 2015 at 4:18 PM, Jonathan Morton <chromatix99@gmail.com> wrote:

> On 30 Mar, 2015, at 16:56, Pedro Tumusok <pedro.tumusok@gmail.com> wrote:<= br> >
> Also would you mind doing another test and share the link for result, = so he can check closer?

I still have the result link for the earlier one:

=C2=A0 =C2=A0 =C2=A0 =C2=A0 http://www.dslreports.com/speedtest/193440
Here=E2=80=99s another result without using the webcache, but keeping every= thing else the same:

=C2=A0 =C2=A0 =C2=A0 =C2=A0 http://www.dslreports.com/speedtest/199524
For context, this is through the PowerBook which is running cake, 8Mbps dow= n and 1Mbps up.=C2=A0 This is somewhat less than the capacity of the link, = but since it=E2=80=99s 3G it varies quite a lot anyway.

> on the other things, yes I'd like to hear about buffer bloat ideas= .

The number-one desirable feature is to carry on measuring the latenc= y while the throughput test is ongoing.=C2=A0 This is relevant to anyone wh= o wants to make a VoIP call or play an online game while using the connecti= on for something else.=C2=A0 If you have several people in one household, i= t=E2=80=99s often hard to coordinate their activity to avoid such multitask= ing.

=C2=A0- Jonathan Morton


I will forward it to him, = hopefully he will signup here also, so the smart people can talk to each ot= her and not having me mess it up :)


--
B= est regards / Mvh
Jan Pedro Tumusok




--
Best regards / Mvh
Jan Pedro Tumusok

--089e0149397afc3e5605128b8bc8--