<br><br>I have to admit that most of the first 20 minutes is review for people here, but I'm hoping that we finally came up with a clear explanation of two of the phases in the codel algorithm, which starts about 28 minutes in. Kathie Nichols, Eric Dumazet and Luigi Rizzo were also there and we got a little sidetracked in discussing some of the interesting results of new sims, and some new lines of research we were exploring, but in viewing the video I'm pretty happy about multiple things. <br>
<br>Video and slides:<br><br><a href="http://www.youtube.com/watch?v=Mxoa5Si4Ubw" target="_blank">http://www.youtube.com/watch?v=Mxoa5Si4Ubw</a><br><br><a href="http://netseminar.stanford.edu/seminars/Inside_Codel_and_Fq_Codel.pdf" target="_blank">http://netseminar.stanford.edu/seminars/Inside_Codel_and_Fq_Codel.pdf</a><br>
<br>The only thing I deeply regret was running out of time and not being able to return to the core slide and talk about the interrelationship of codel's drop scheduler with a fair queueing algorithm. I spent too much time showing off some cool graphs of the results I currently get from the RRUL test prototype.<br>
<br>Share and enjoy!<br><br>Other lectures from the Stanford netseminar series are here:<br><br><a href="http://netseminar.stanford.edu/" target="_blank">http://netseminar.stanford.edu/</a><br><br>I have a talk at MIT next week on the 14th.<br>
<br><br>-- <br>Dave Täht<br><br>Fixing bufferbloat with cerowrt: <a href="http://www.teklibre.com/cerowrt/subscribe.html" target="_blank">http://www.teklibre.com/cerowrt/subscribe.html</a>