<html><head><meta http-equiv="Content-Type" content="text/html charset=utf-8"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class=""><br class=""><div><blockquote type="cite" class=""><div class="">On Apr 16, 2018, at 11:23 PM, Toke Høiland-Jørgensen <<a href="mailto:toke@toke.dk" class="">toke@toke.dk</a>> wrote:</div><div class=""><div class=""><br class=""><blockquote type="cite" class="">I remember that fairness behavior at low RTTs (< 20ms) needed to be<br class="">either improved or documented, and don’t see anything about that in<br class="">the man page in the tc-adv repo thus far. Summarizing the host<br class="">isolation results at<br class=""><a href="http://www.drhleny.cz/bufferbloat/cake/round2/#hostiso_cake_{rtt}_{qos-id}" class="">http://www.drhleny.cz/bufferbloat/cake/round2/#hostiso_cake_{rtt}_{qos-id}</a><br class=""><http://www.drhleny.cz/bufferbloat/cake/round2/#hostiso_cake_%7Brtt%7D_%7Bqos-id%7D>:<br class=""><br class="">RTT: fairness (1.0 == perfect fairness)<br class="">---<br class="">100us: 2.22<br class="">1ms: 1.7<br class="">2ms: 1.6<br class="">3ms: 1.42<br class="">5ms: 1.31<br class="">8ms: 1.16<br class="">10ms: 1.12<br class="">20ms: 1.02<br class="">40ms: 1.017<br class=""></blockquote><br class="">Erm, what's the metric and which data source are you looking at here?<br class=""></div></div></blockquote></div><div class=""><br class=""></div><div class="">Subject changed...</div><br class=""><div class=""><div class="">The clients were as follows:</div><div class="">Client 0- 1 stream up</div><div class="">Client 1- 12 streams up</div><div class="">Client 2- 1 stream down</div><div class="">Client 3- 12 streams down</div></div><div class=""><br class=""></div><div class="">It looks like what I used before was Client 3’s “TCP Download Sum avg” divided by Client 2’s “TCP Download avg” from the srchost/dsthost tests. The data’s in a table here (see column “Client 3 Mean / Client 2 Mean”):</div><div class=""><br class=""></div><div class=""><a href="https://docs.google.com/spreadsheets/d/1e06ZfHSSmecJx9sPU2s2g2GYCS18cMIhBN8PXf1jwaM/edit?usp=sharing" class="">https://docs.google.com/spreadsheets/d/1e06ZfHSSmecJx9sPU2s2g2GYCS18cMIhBN8PXf1jwaM/edit?usp=sharing</a></div><div class=""><br class=""></div><div class="">I was calculating by hand before, so the numbers are slightly different, but that’s the idea.</div><div class=""><br class=""></div><div class="">Now, these tests were done at 500Mbit between two cabled APU2s, so we could just be running out of CPU on this hardware at lower RTTs. There are CPU stats included, and a “Flent Data Files” section. Is it possible to tell from this if CPU is the problem? The highest median client load I see looks to be 0.77 for the 40ms tests, for example, with a mean of 0.63.</div><div class=""><br class=""></div><div class=""><div class=""><a href="http://www.drhleny.cz/bufferbloat/cake/round2/hostiso_cake_100us_eg_cakeeth_src_cakeeth_dst_500mbit/index.html" class="">http://www.drhleny.cz/bufferbloat/cake/round2/hostiso_cake_100us_eg_cakeeth_src_cakeeth_dst_500mbit/index.html</a></div><div class=""><br class=""></div><div class=""><a href="http://www.drhleny.cz/bufferbloat/cake/round2/hostiso_cake_40ms_eg_cakeeth_src_cakeeth_dst_500mbit/index.html" class="">http://www.drhleny.cz/bufferbloat/cake/round2/hostiso_cake_40ms_eg_cakeeth_src_cakeeth_dst_500mbit/index.html</a></div></div><div class=""><br class=""></div><div class="">If we’re not sure, the tests would have to be redone at a lower bitrate. It would probably be best if someone reproduced these results externally anyway, and perhaps it would also be better to not be mixing upload and download tests at the same time… :)</div><div class=""><br class=""></div></body></html>