Re: [Bloat] Netflix's fast.com now measuring bufferbloat

2018-07-20 Thread Michael Richardson
Oh, it's also not obvious that it tests IPv6 if it's there. So sitting in a pub, it tested the speed of my IPv6 VPN, not the pub's speed :-) Arie wrote: > In addition to now measuring upload speed and idle latency, it also > measures latency under load. The blog post announcing the

Re: [Bloat] Netflix's fast.com now measuring bufferbloat

2018-07-20 Thread Michael Richardson
Arie wrote: > In addition to now measuring upload speed and idle latency, it also > measures latency under load. The blog post announcing the change also > has a shout out to bufferbloat.net > It's nice having another option than just DSLReports to point people > to when they

Re: [Bloat] [Make-wifi-fast] Van Jacobson's slides on timing wheels at netdevconf

2018-07-20 Thread Luca Muscariello
and C) you can implement any packet scheduler using a timing wheel using virtual times. On Fri, Jul 20, 2018 at 4:09 PM Dave Taht wrote: > > https://www.files.netdevconf.org/d/46def75c2ef345809bbe/files/?p=/Evolving%20from%20AFAP%20%E2%80%93%20Teaching%20NICs%20about%20time.pdf > > Talking

[Bloat] Van Jacobson's slides on timing wheels at netdevconf

2018-07-20 Thread Dave Taht
https://www.files.netdevconf.org/d/46def75c2ef345809bbe/files/?p=/Evolving%20from%20AFAP%20%E2%80%93%20Teaching%20NICs%20about%20time.pdf Talking about replacing queues with timing wheels. I will refrain from commenting other than noting A) I like it. We essentially have limits in the OS on