On Fri, Feb 27, 2015 at 10:26 AM, Ronald Pina <pinaronald@xxxxxxxxx> wrote: > It obvious that fixing bufferbloat problems has made great steps > toward, the throughput overhead of fq_codel is a at low levels and can > be negligible compared to the its advantages , I can submit some chart > to this topic as soon as i get some free time... > > Regards Awesome. More independent benchmarks and testing is really needed, by people new to the concepts. I appreciate whatever you can do. One more - fun - item, is to try enabling ecn on your tcps and running that through fq_codel (which has ecn enabled by default). That is a single sysctl on both sides of the connection: net.ipv4.tcp_ecn=1 that hits the trifecta - low latency, high throughput, and zero packet loss. We just have a few billion machines to upgrade. > > On Fri, Feb 27, 2015 at 6:55 PM, Dave Taht <dave.taht@xxxxxxxxx> wrote: >> On Fri, Feb 27, 2015 at 9:52 AM, Ronald Pina <pinaronald@xxxxxxxxx> wrote: >>> I have been tested DRR using fq_codel and codel in an extremely >>> congested network and I can confirm that the average packet latency >>> has been improved quite well comparing to fifo, the latency is keep at >>> low levels and the throughput tends to be stable. > >> Isn't it awesome? We can replace the phone system now... >> >> I try to encourage people to use netperf-wrapper to give easy >> comparison charts from repeatable tests. Did you try that? -- Dave Täht Let's make wifi fast, less jittery and reliable again! https://plus.google.com/u/0/107942175615993706558/posts/TVX3o84jjmb -- To unsubscribe from this list: send the line "unsubscribe lartc" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
