<div dir="ltr">Dave,<div>What is the highest bandwidth that you have been able to shape with CAKE on the APU2?</div><div><br></div><div>Regards,</div><div>Luis Garcia</div></div><div class="gmail_extra"><br><div class="gmail_quote">On Wed, Dec 6, 2017 at 3:08 PM, Dave Taht <span dir="ltr"><<a href="mailto:dave@taht.net" target="_blank">dave@taht.net</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><br>
Groovy.<br>
<br>
I guess if I had any one goal, was that I wanted to have cake to shape<br>
well on the APU2 at a gigabit. That's the highest end "home router" we<br>
have, although it would be good to get results on things like the turris<br>
omnia, also.<br>
<br>
I hope to start a round of profiling as to what can be optimized (out)<br>
this weekend. Last I looked the hashing costs dominated.<br>
<br>
Pretty sure we're good at 200Mbit there (aside from the ingress bug).<br>
<br>
It would be nice to have a test showing blue being useful.<br>
<br>
Did you get cpu stats for the below tests?<br>
<br>
Georgios Amanakis <<a href="mailto:gamanakis@gmail.com">gamanakis@gmail.com</a>> writes:<br>
<br>
> -------- Forwarded Message --------<br>
> From: Georgios Amanakis <<a href="mailto:gamanakis@gmail.com">gamanakis@gmail.com</a>><br>
> To: Dave Taht <<a href="mailto:dave.taht@gmail.com">dave.taht@gmail.com</a>><br>
> Cc: Cake List <<a href="mailto:cake@lists.bufferbloat.net">cake@lists.bufferbloat.net</a>>, Jonathan Morton <chromatix9<br>
> <a href="mailto:9@gmail.com">9@gmail.com</a>>, Pete Heist <<a href="mailto:peteheist@gmail.com">peteheist@gmail.com</a>><br>
> Subject: Re: [Cake] RHODIUM - nuking blue (for testing)<br>
> Date: Tue, 05 Dec 2017 20:58:39 -0500<br>
><br>
> I just finished testing Rhodium, and I am supplementing Sunday's<br>
> results.<br>
><br>
> CMTS setup:<br>
> server -- delay -- isp -- mbox -- client<br>
><br>
> ISP limited at 200/10mbit.<br>
> Cake/HTB shaping at 180/9mbit.<br>
> RTT 20ms with cake's RTT set at 20ms<br>
> rrul_be_nflows, 32 download, 8 upload streams.<br>
> net.ipv4.tcp_ecn=2 in all nodes (linux default)<br>
><br>
> Cake tested in all flow-isolation modes with all ack filtering options.<br>
> HTB tested with codel,pie,fq,sfq,fq_codel.<br>
><br>
> Rhodium(cakerhod) doesn't show much difference against Cobalt(cake).<br>
><br>
> George<br>
><br>
> On Tue, 2017-12-05 at 11:32 -0800, Dave Taht wrote:<br>
>> Given some of the results thus far at various RTTs, and at higher<br>
>> loads, like the ginormous one georgios just ran, I thought it might<br>
>> be<br>
>> useful to run a battery of tests with blue disabled.<br>
>><br>
>> I haven't gone to the trouble of creating a formal branch - tho if I<br>
>> did, I'd call it rhodium, as rhodium detectors are used in nuclear<br>
>> reactors to measure the neutron flux level.<br>
>><br>
>> untested-but-compiling-patch attached. I'm busy on a few other things<br>
>> this week, and it is looking like we'll have to replace the xstats<br>
>> structure with something smaller to go mainline.<br>
>><br>
>> ______________________________<wbr>_________________<br>
>> Cake mailing list<br>
>> <a href="mailto:Cake@lists.bufferbloat.net">Cake@lists.bufferbloat.net</a><br>
>> <a href="https://lists.bufferbloat.net/listinfo/cake" rel="noreferrer" target="_blank">https://lists.bufferbloat.net/<wbr>listinfo/cake</a><br>
><br>
><br>
><br>
><br>
> ______________________________<wbr>_________________<br>
> Cake mailing list<br>
> <a href="mailto:Cake@lists.bufferbloat.net">Cake@lists.bufferbloat.net</a><br>
> <a href="https://lists.bufferbloat.net/listinfo/cake" rel="noreferrer" target="_blank">https://lists.bufferbloat.net/<wbr>listinfo/cake</a><br>
______________________________<wbr>_________________<br>
Cake mailing list<br>
<a href="mailto:Cake@lists.bufferbloat.net">Cake@lists.bufferbloat.net</a><br>
<a href="https://lists.bufferbloat.net/listinfo/cake" rel="noreferrer" target="_blank">https://lists.bufferbloat.net/<wbr>listinfo/cake</a><br>
</blockquote></div><br></div>