[Starlink] fiber IXPs in space

[Ulrich Speidel]

For all I can tell, dishy only uses one satellite at a time and so 
routing via dishys to other satellites - while certainly an attractive 
concept - doesn't seem to be happening.

Most modern comms systems - look at your phone - run pretty complicated 
stacks these days, and I'd be very surprised if Starlink didn't. That 
makes it likely that there is an extra network and transport layer 
between dishy and gateway (a term, which I'm incidentally not using for 
the WiFi router that comes with dishy but for the gateways that connect 
the Starlink space network to the terrestrial Internet). Any routing 
between dishy, satellite and gateway (and any lasers along the way) will 
therefore be happening at that extra network layer - the transport layer 
above it has the job to make the path between dishy and gateway look 
like a single data link layer hop. This is why you don't see a satellite 
IP in your traceroute. Quite how that extra network and transport layer 
is implemented is something that probably only Starlink knows. The power 
of layered communication ;-)

P.S.: I'll apologise in advance for upcoming silence over the next few 
days - we're taking Dishy for an outing to get it away from the Clevedon 
gateway, so we can finally get Dave some more data ;-)

On 18/04/2023 7:09 am, David Lang via Starlink wrote:
> if Starlink can route via in-space-lasers and in a dishy-to-dishy way 
> (both have
> been talked about, at least in future tenses) then they could also 
> route to an
> on-satellite IP.
>
> historically 'bent pipe' satellite support meant that the satellite just
> repeated the RF signal back down with no modifications. Starlink was 
> designed to
> do routing of traffic, some to a ground station (possibly more than 
> one), some
> to other satellites (including ones at different altitudes), and some 
> to other
> dishys. It's initial deployment included no routing, just relaying 
> between a
> dishy and a ground station, but we know that it's extended beyond 
> that, at least
> when there are not ground stations in range
>
> David Lang
>
> On Mon, 17 Apr 2023, David Fernández via Starlink wrote:
>
> > Well, I have some concerns about how you implement an anycast address
> > in a transparent satellite.
> >
> > If the pre-requisite for this is that the satellite is a router, I
> > don't see this happening anytime soon. I am not aware of any system,
> > not deployed, even designed with satellites being routers, but IRIS2
> > could be the first, maybe:
> > 
> https://defence-industry-space.ec.europa.eu/eu-space-policy/eu-space-programme/iriss_en 
> <https://defence-industry-space.ec.europa.eu/eu-space-policy/eu-space-programme/iriss_en>
> >
> > Bufferbloat will be checked and prevented as much as possible in IRIS2.
> >
> > Regards,
> >
> > David
> >
> > 2023-04-17 16:38 GMT+02:00, Rodney W. Grimes 
> <starlink at gndrsh.dnsmgr.net>:
> >>> On Sun, 16 Apr 2023, David Fern?ndez via Starlink wrote:
> >>>
> >>> > The idea would be that the satellite inspects IP packets and when it
> >>> > detects a DNS query, instead of forwarding the packet to ground
> >>> > station, it just answers back to the sender of the query.
> >>>
> >>> This would be a bad way to implement it. You don't want to override
> >>> queries to
> >>> other DNS servers, but it would be very easy to create an anycast 
> address
> >>> that
> >>> is served by the satellites.
> >>
> >> Yes, and the later is what I proposed, the idea of intercepting
> >> someone ELSE'S anycast address and processing it would be
> >> wrong in many ways, in effect a Man In the Middle attack
> >> as stated else where.
> >>
> >>> David Lang
> >> --
> >> Rod Grimes
> >> rgrimes at freebsd.org
> >>
> > _______________________________________________
> > Starlink mailing list
> > Starlink at lists.bufferbloat.net
> > https://lists.bufferbloat.net/listinfo/starlink 
> <https://lists.bufferbloat.net/listinfo/starlink> 
>
>
> _______________________________________________
> Starlink mailing list
> Starlink at lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/starlink
>
-- 
****************************************************************
Dr. Ulrich Speidel

School of Computer Science

Room 303S.594 (City Campus)

The University of Auckland
u.speidel at auckland.ac.nz  
http://www.cs.auckland.ac.nz/~ulrich/
****************************************************************


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.bufferbloat.net/pipermail/starlink/attachments/20230418/d6da44b3/attachment-0001.html>