From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-oa0-x22e.google.com (mail-oa0-x22e.google.com [IPv6:2607:f8b0:4003:c02::22e]) (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 4284921F6AA for ; Wed, 6 Aug 2014 16:10:51 -0700 (PDT) Received: by mail-oa0-f46.google.com with SMTP id m1so2358770oag.5 for ; Wed, 06 Aug 2014 16:10:50 -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=vlHfBidkiVIGaCMXwyRrs4k1SZKRgSY6v3cKo8TiwrU=; b=RZNu0sCtmOr2KXZYZ+K4XmCN3bM37yWnGKAykY2DBdzD2qAqzVeVtGEg9/+E7Rx39P PzDG+jxkMB6kzc6eknjZDhGqgqy7xX/Xgm8vPImQNp8IvKvT+SV5g7UgM1SsYbr+K7Tb pi5cNSCsd+YYy/ZTWPrl1Q5Z0KuPDyqNIqTYax+FPqONtw8O09sP3YTRVsrBnw7jkwpX hLNKTGpuOCjuInbCaTk6SjYQOkjhxVx3HzUdkvLiOF3fRhsx2fyA3EKVcxXyXmZoAO/A YRS19rUGyozJI7xoLcpI7BjNtJfUPWEOG6J6yxqfosaVRSreiSVbNJ5lM27i4/j8coJB aILQ== MIME-Version: 1.0 X-Received: by 10.60.23.39 with SMTP id j7mr19597688oef.20.1407366650426; Wed, 06 Aug 2014 16:10:50 -0700 (PDT) Received: by 10.202.93.69 with HTTP; Wed, 6 Aug 2014 16:10:50 -0700 (PDT) Received: by 10.202.93.69 with HTTP; Wed, 6 Aug 2014 16:10:50 -0700 (PDT) In-Reply-To: References: <53BEA813.8000108@gmail.com> <53E2839D.8090304@gmail.com> Date: Wed, 6 Aug 2014 19:10:50 -0400 Message-ID: From: Dave Taht To: David Lang Content-Type: multipart/alternative; boundary=047d7b33d500447c0304fffe15f9 Cc: cerowrt-devel@lists.bufferbloat.net Subject: Re: [Cerowrt-devel] Upper routing throughput limit X-BeenThere: cerowrt-devel@lists.bufferbloat.net X-Mailman-Version: 2.1.13 Precedence: list List-Id: Development issues regarding the cerowrt test router project List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 06 Aug 2014 23:10:51 -0000 --047d7b33d500447c0304fffe15f9 Content-Type: text/plain; charset=UTF-8 On Aug 6, 2014 4:06 PM, "David Lang" wrote: > > When link speeds get high enough, do we still need to shape the download for home users? at some point you stop saturating the line. > > the 3800 can easily handle 100mb if it's not trying to shape the traffic, is it worth seeing if there's any way to squeeze that shaping overhead down? In other words if you set the download to 0 in sqm, that disables ingress shaping completely... and cero can handle the 35mbit upload without raising too much sweat I have a report of a 250 mbit download cable modem exhibiting 300ms of latency. That would be both sad and good to confirm... > > David Lang > > On Wed, 6 Aug 2014, William Katsak wrote: > >> Date: Wed, 06 Aug 2014 15:35:57 -0400 >> From: William Katsak >> To: "cerowrt-devel@lists.bufferbloat.net" >> >> Subject: Re: [Cerowrt-devel] Upper routing throughput limit >> >> >> Sorry to wake this thread up again. It does indeed seem that the 3800 is having trouble routing at my line speed. >> >> I was considering doing a pfSense box, but it doesn't seem that Bufferbloat has been much of a consideration yet over there. There is a version of Codel, but the QoS would have to be set up manually. >> >> I've Googled this Ubiquiti Edgerouter Lite, and I am intrigued. I don't see many details on Ubiquiti's site about the QoS though. Is this device as good at beating bloat as Cero? Would mating one of these with a 3800 (for Wifi only) be a good bet? >> >> Thanks, >> -Bill >> >> >> On 07/10/2014 11:01 AM, Aaron Wood wrote: >>> >>> It depends on the aqm rules that are configured. In the base setup, it >>> struggles at 50Mbps. But that can be increased by switching from the >>> simple.qos script to simplest.qos (I'm not sure where the limit is with >>> the simplest.qos script. >>> >>> I know that Dave Taht has been working with some other platforms. The >>> Ubiquity EdgeRouter Lite may be able to hit 100Mbps, but it doesn't run >>> CeroWRT itself, it just supports similar configuration. >>> >>> -Aaron >>> >>> >>> On Thu, Jul 10, 2014 at 7:49 AM, William Katsak >> > wrote: >>> >>> Hello, >>> >>> Does anyone have a good sense of the most throughput our 3800s with >>> Cero can push through the WAN interface? Are we good to 100 mbps? 1gbps? >>> >>> Thanks, >>> Bill Katsak >>> _________________________________________________ >>> Cerowrt-devel mailing list >>> Cerowrt-devel@lists.__bufferbloat.net >>> >>> https://lists.bufferbloat.net/__listinfo/cerowrt-devel >>> >>> >>> >> _______________________________________________ >> Cerowrt-devel mailing list >> Cerowrt-devel@lists.bufferbloat.net >> https://lists.bufferbloat.net/listinfo/cerowrt-devel >> > _______________________________________________ > Cerowrt-devel mailing list > Cerowrt-devel@lists.bufferbloat.net > https://lists.bufferbloat.net/listinfo/cerowrt-devel --047d7b33d500447c0304fffe15f9 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable


On Aug 6, 2014 4:06 PM, "David Lang" <david@lang.hm> wrote:
>
> When link speeds get high enough, do we still need to shape the downlo= ad for home users? at some point you stop saturating the line.
>
> the 3800 can easily handle 100mb if it's not trying to shape the t= raffic, is it worth seeing if there's any way to squeeze that shaping o= verhead down?

In other words if you set the download to 0 in sqm, that dis= ables ingress shaping completely... and cero can handle the 35mbit upload w= ithout raising too much sweat

I have a report of a 250 mbit download cable modem exhibitin= g 300ms of latency. That would be both sad and good to confirm...

>
> David Lang
>
> On Wed, 6 Aug 2014, William Katsak wrote:
>
>> Date: Wed, 06 Aug 2014 15:35:57 -0400
>> From: William Katsak <wkat= sak@gmail.com>
>> To: "c= erowrt-devel@lists.bufferbloat.net"
>> =C2=A0 =C2=A0 <cerowrt-devel@lists.bufferbloat.net>
>> Subject: Re: [Cerowrt-devel] Upper routing throughput limit
>>
>>
>> Sorry to wake this thread up again. It does indeed seem that the 3= 800 is having trouble routing at my line speed.
>>
>> I was considering doing a pfSense box, but it doesn't seem tha= t Bufferbloat has been much of a consideration yet over there. There is a v= ersion of Codel, but the QoS would have to be set up manually.
>>
>> I've Googled this Ubiquiti Edgerouter Lite, and I am intrigued= . I don't see many details on Ubiquiti's site about the QoS though.= Is this device as good at beating bloat as Cero? Would mating one of these= with a 3800 (for Wifi only) be a good bet?
>>
>> Thanks,
>> -Bill
>>
>>
>> On 07/10/2014 11:01 AM, Aaron Wood wrote:
>>>
>>> It depends on the aqm rules that are configured. =C2=A0In the = base setup, it
>>> struggles at 50Mbps. =C2=A0But that can be increased by switch= ing from the
>>> simple.qos script to simplest.qos (I'm not sure where the = limit is with
>>> the simplest.qos script.
>>>
>>> I know that Dave Taht has been working with some other platfor= ms. =C2=A0The
>>> Ubiquity EdgeRouter Lite may be able to hit 100Mbps, but it do= esn't run
>>> CeroWRT itself, it just supports similar configuration.
>>>
>>> -Aaron
>>>
>>>
>>> On Thu, Jul 10, 2014 at 7:49 AM, William Katsak <wkatsak@gmail.com
>>> <mailto:wkatsak@gmail.= com>> wrote:
>>>
>>> =C2=A0 =C2=A0 Hello,
>>>
>>> =C2=A0 =C2=A0 Does anyone have a good sense of the most throug= hput our 3800s with
>>> =C2=A0 =C2=A0 Cero can push through the WAN interface? Are we = good to 100 mbps? 1gbps?
>>>
>>> =C2=A0 =C2=A0 Thanks,
>>> =C2=A0 =C2=A0 Bill Katsak
>>> =C2=A0 =C2=A0 ________________________________________________= _
>>> =C2=A0 =C2=A0 Cerowrt-devel mailing list
>>> =C2=A0 =C2=A0 Cerowrt-devel@lists.__bufferbloat.net
>>> =C2=A0 =C2=A0 <mailto:Cerowrt-devel@lists.bufferbloat.net>
>>> =C2=A0 =C2=A0 https://lists.bufferbloat.net/__listinfo/cerowrt-devel
>>> =C2=A0 =C2=A0 <https://lists.bufferbloat.net/listinfo/cerowrt-devel>
>>>
>>>
>> _______________________________________________
>> Cerowrt-devel mailing list
>> Cerowrt-dev= el@lists.bufferbloat.net
>> h= ttps://lists.bufferbloat.net/listinfo/cerowrt-devel
>>
> _______________________________________________
> Cerowrt-devel mailing list
> Cerowrt-devel@l= ists.bufferbloat.net
> https= ://lists.bufferbloat.net/listinfo/cerowrt-devel

--047d7b33d500447c0304fffe15f9--