[Starlink] keep the pipe full but no fuller

2024-02-18 Thread Dave Taht via Starlink
This is a nice analysis and a few animations of some core queue theory, tying back to Len Kleinrock's paper. https://dipsingh.github.io/Keep-The-Pipes-Full/ -- 40 years of net history, a couple songs: https://www.youtube.com/watch?v=D9RGX6QFm5E Dave Täht CSO, LibreQos ___

Re: [Starlink] Starlink REV4 overview and tests

2024-02-18 Thread Oleg Kutkov via Starlink
It seems that the latest firmware has some improvements: https://i0.wp.com/olegkutkov.me/wp-content/uploads/2024/02/rrul_-_rev4_5776ed0f-38b5-47bf-8ff4-d38f7d2c326b_release.png?ssl=1 300 seconds test: https://olegkutkov.me/starlink_rev4_second_test/rrul-2024-02-18T203347.764414.rev4_5776ed0f-38

Re: [Starlink] Starlink REV4 overview and tests

2024-02-18 Thread Dave Taht via Starlink
The smoothness that you are observing, may be an artifact of the sample rate you ran the tests at, which appears to be 1/2 second. --step-size=.05 is a much better sample rate. On Sun, Feb 18, 2024 at 4:47 PM Oleg Kutkov via Starlink wrote: > > It seems that the latest firmware has some improve

Re: [Starlink] Starlink REV4 overview and tests

2024-02-18 Thread Oleg Kutkov via Starlink
https://pbs.twimg.com/media/GGqGfyGX0AEbl3S?format=jpg&name=large https://pbs.twimg.com/media/GGqGfxhWYAAhj4t?format=jpg&name=large On 2/19/24 00:25, Dave Taht wrote: The smoothness that you are observing, may be an artifact of the sample rate you ran the tests at, which appears to be 1/2 secon

Re: [Starlink] keep the pipe full but no fuller

2024-02-18 Thread Hesham ElBakoury via Starlink
How I can find the animation? Hesham On Sun, Feb 18, 2024, 4:13 AM Dave Taht via Starlink < starlink@lists.bufferbloat.net> wrote: > This is a nice analysis and a few animations of some core queue > theory, tying back to Len Kleinrock's paper. > > https://dipsingh.github.io/Keep-The-Pipes-Full/