[Bloat] FQ-PIE kernel module implementation

Hi all, I’m Hironori Okano and Fred’s intern. I’d like to let you know that I have implemented FQ-PIE as a linux kernel module “fq-pie and iproute2 for fq-pie. This was done in collaboration with others at Cisco including Fred Baker, Rong Pan, Bill Ver Steeg, and Preethi Natarajan. The source

Re: [Bloat] Bloat goes away, but with ~25% speed loss?

What about the link type? If there are extra overheads going on, that's going to muck with the calculations (possibly adding latency, but shouldn't be cutting bandwidth), since the throttling calculations will be wrong. His ISP may be able to help with that. It would be interesting to see what

Re: [Bloat] I thought the diagram, at least, was a good start

On ubuntu 14.04, I add a file named 10-debloat.conf in /etc/sysctl.d with the following lines in it: jg@thinkpad:/etc/sysctl.d$ cat 10-debloat.conf net.ipv4.tcp_ecn=1 net.core.default_qdisc=fq_codel On Thu, Jun 4, 2015 at 4:35 PM, Jonathan Morton chromati...@gmail.com wrote: The main one on

Re: [Bloat] I thought the diagram, at least, was a good start

The main one on Linux would be: echo net.core.default_qdisc=fq_codel /etc/sysctl.conf ...at least, if I've remembered it right. - Jonathan Morton ___ Bloat mailing list Bloat@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] Bloat goes away, but with ~25% speed loss?

I think he may be seeing a complex interaction of various different queue components and bottlenecks, which in total manages to confuse TCP congestion control sufficiently to cause reduced throughput. I suspect that there is a shaper at the ISP end which limits the bandwidth available to any