Starlink has bufferbloat. Bad.
 help / color / mirror / Atom feed
From: Dave Taht <davet@teklibre.net>
To: Mikael Abrahamsson <swmike@swm.pp.se>
Cc: David Lang <david@lang.hm>, starlink@lists.bufferbloat.net
Subject: Re: [Starlink] 69,000 Users
Date: Wed, 30 Jun 2021 07:24:53 -0700	[thread overview]
Message-ID: <5E2A258B-4C5F-476E-A61D-1267993C50FB@teklibre.net> (raw)
In-Reply-To: <alpine.DEB.2.20.2106301151280.13354@uplift.swm.pp.se>



> On Jun 30, 2021, at 2:57 AM, Mikael Abrahamsson <swmike@swm.pp.se> wrote:
> 
> On Wed, 30 Jun 2021, David Lang wrote:
> 
>> I suspect that they will be more limited by the number of stations they can build than the interest from customers. As user density increases, they will need to launch more satellites, but as Starship comes online, the cost to do so will drop significantly.
> 
> My opinion here is that it's a bw game. with 1M customers doing a few megabits/s each, that's significant amount of bw capacity needed. Even with tens of thousands of satellites (each significant cost to build and launch), I still don't really see how they'll handle many millions of customers.
> 
> Terrestrial mobile networks still haven't really come around to unlimited data that actually works, so we'll see how well Starlink can do this.
> 

Terrestrial mobile networks still haven’t implemented fq and aqm technologies. 

> There was a power outage affecting around 50k households yesterday, I presume most peoples' residential connections/wifis stopped working, and they all went mobile. This rendered the local mobile networks basically unusable, people reported 10s RTT on some of the packets that were actually delivered.

Link?

At 250+ms most our protocols start sending more packets, compounding this problem.

Admittedly an outage of this size is probably not something that can be handled with pure fq+aqm, as it would basically look more like a syn flood attack!


> 5G base stations today are in the gigabits/s magnitude of total air capacity, even if they have similar for the satellites it's going to be problematic to keep up. I think they'll have to keep charging a premium price, higher than today, or implement data cap.

I don’t think data caps are needed. fairness is needed.

I would prefer a solution that just billed for usage over a minimum. 

> 
> -- 
> Mikael Abrahamsson    email: swmike@swm.pp.se
> _______________________________________________
> Starlink mailing list
> Starlink@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/starlink


  reply	other threads:[~2021-06-30 14:24 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-30  2:00 Daniel AJ Sokolov
2021-06-30  5:24 ` David Lang
2021-06-30  5:48   ` Dave Taht
2021-06-30  7:13     ` [Starlink] routing capability in starlinks Dave Taht
2021-06-30 14:43       ` Michael Richardson
2021-06-30 23:56         ` Nick Buraglio
2021-06-30  7:23     ` [Starlink] 69,000 Users Mike Puchol
2021-06-30  7:30       ` David Lang
2021-06-30  7:43         ` Mike Puchol
2021-06-30  7:51         ` Dave Taht
2021-06-30  9:57         ` Mikael Abrahamsson
2021-06-30 14:24           ` Dave Taht [this message]
2021-06-30 18:33             ` Mikael Abrahamsson
2021-06-30 20:40               ` Dick Roy
2021-06-30 23:33         ` Daniel AJ Sokolov
2021-07-01  0:00           ` David Lang
2021-07-01  0:03             ` Daniel AJ Sokolov
2021-07-01  0:20               ` David Lang
2021-06-30 10:53       ` Jared Mauch

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: https://lists.bufferbloat.net/postorius/lists/starlink.lists.bufferbloat.net/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=5E2A258B-4C5F-476E-A61D-1267993C50FB@teklibre.net \
    --to=davet@teklibre.net \
    --cc=david@lang.hm \
    --cc=starlink@lists.bufferbloat.net \
    --cc=swmike@swm.pp.se \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox