Re: [Starlink] musk: 28ms median latency on starlink

[David Lang <david@lang.hm>]

[-- Attachment #1: Type: text/plain, Size: 2995 bytes --]

On Tue, 4 Jun 2024, Alexandre Petrescu via Starlink wrote:

> Le 02/06/2024 à 19:13, Dave Taht via Starlink a écrit :
>> Via elon musk:
>> 
>> Starlink just achieved a new internal median latency record of 28ms 
>> yesterday! Great work by the engineering and operations teams.
>
> It is indeed excellent news to hear that they achieve such a low latency.  It 
> is worth noting.
>
> In addition to fq_codel, cake and libreqos (acronyms I simply cite), if I may 
> speculate how to further decrease the latency:
>
> - lower the sat altitudes, maybe at 350km or even at 70km. These are 
> altitudes where starlink sats seem parked at times, maybe right before 
> reaching their stable 500km altitudes.  It means they can stay there, maybe 
> for a shorter time than the times they stay at 500km, but still, for some 
> time.

they do have orbital shells scheduled for the 350km range

> - make the antennas on the sats more 'wide-angle' so to speak, such as to 
> make multiple spots on the ground, possible very distanced, rather than one 
> single spot straight below the sat.

we know that they have several antennas on each sat, and the V2 sats will have 
more (and probably bigger ones)

> - add sat-to-HAP links. (high-alti platforms, i.e. 20km-to-70km), 
> sat-to-plane, sat-to drone links (altitudes 500m-20km).

would those really be much different than the existing ground stations?

> - add sat-to-other-constellations' sats links.

well, there need to be other constellations to connect to first :-)

> - have a dynamic IP routing system between sats and HAPS naturally finding 
> shortest paths, i.e. lowest latencies.
>
> I think one could target a 1ms latency.

even down at 300Km, and assuming the satellite is directly overhead, that's 
~5ms round trip, and to avoid interference with higher satellites, they don't 
operate directly overhead, I think 10ms is more in the ballpark of theoretical 
maximum, and a goal of 20ms is aggressive.

David Lang

> Alex
>
>> 
>> - https://twitter.com/elonmusk/status/1797282250574184587
>> 
>> I of course, am very interested in y'all´s external measurements of how 
>> well starlink is doing. For me, it is fantastic - 30Mbit uploads nowadays, 
>> 0
>> latency on the upload (how?) 
>> https://www.waveform.com/tools/bufferbloat?test-id=2a1d139b-87cb-4ba4-a829-e2167801cffe
>> 
>> I also keep hoping that the rest of the ISP industry is now paying 
>> attention and deploying stuff like fq_codel and cake and libreqos, but, ah 
>> well - I will settle for starlink blowing past a lot of dsl and cable and 
>> finding ways to get their density up.
>> 
>> Anyone going to the Starship launch on the 6th?
>> 
>> 
>> 
>> -- 
>> https://www.youtube.com/watch?v=BVFWSyMp3xg&t=1098s 
>> <https://www.youtube.com/watch?v=BVFWSyMp3xg&t=1098s> Waves Podcast
>> Dave Täht CSO, LibreQos
>> 
>> _______________________________________________
>> Starlink mailing list
>> Starlink@lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/starlink

[-- Attachment #2: Type: text/plain, Size: 149 bytes --]

_______________________________________________
Starlink mailing list
Starlink@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/starlink