From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from smtp118.iad3a.emailsrvr.com (smtp118.iad3a.emailsrvr.com [173.203.187.118]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id 33BAE3B2A4 for ; Wed, 31 Aug 2022 16:24:22 -0400 (EDT) Received: from app29.wa-webapps.iad3a (relay-webapps.rsapps.net [172.27.255.140]) by smtp31.relay.iad3a.emailsrvr.com (SMTP Server) with ESMTP id C9BCE239EE for ; Wed, 31 Aug 2022 16:24:21 -0400 (EDT) Received: from deepplum.com (localhost.localdomain [127.0.0.1]) by app29.wa-webapps.iad3a (Postfix) with ESMTP id 8DE5520048 for ; Wed, 31 Aug 2022 16:24:21 -0400 (EDT) Received: by apps.rackspace.com (Authenticated sender: dpreed@deepplum.com, from: dpreed@deepplum.com) with HTTP; Wed, 31 Aug 2022 16:24:21 -0400 (EDT) X-Auth-ID: dpreed@deepplum.com Date: Wed, 31 Aug 2022 16:24:21 -0400 (EDT) From: "David P. Reed" To: starlink@lists.bufferbloat.net MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_20220831162421000000_81250" Importance: Normal X-Priority: 3 (Normal) X-Type: html In-Reply-To: References: X-Client-IP: 209.6.168.128 Message-ID: <1661977461.578310669@apps.rackspace.com> X-Mailer: webmail/19.0.18-RC X-Classification-ID: 8e2b75e5-3a8a-44a6-b818-2e932a357c7e-1-1 Subject: [Starlink] CDNs in space! X-BeenThere: starlink@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: "Starlink has bufferbloat. Bad." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 31 Aug 2022 20:24:22 -0000 ------=_20220831162421000000_81250 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable =0AHaving looked into this a lot, CDNs don't account for very much Internet= traffic. There's a lot of marketing propaganda out there that suggests thi= s might be true, but when you try to track down the source, it's almost alw= ays from an old slide deck presented by a company that sells CDN servers.= =0A =0AThat doesn't mean that CDN servers don't help a fair bit, but no, to= day's Internet wouldn't fall apart if we didn't have CDN servers withint 2 = msec. of the edge.=0A =0AAlso, CDN's need to be BIG to hold all the videos = that people might choose to watch at any particular time.=0A =0ASo I'm just= pointing out that the business case for CDN's in space to merely solve Sta= rlink's potential issues is probably not great. Maybe 10 years from now. Ma= ybe never. The idea that everyone watches TV and the same few seconds of co= ntent of a few shows that are extraordinarily popular - well, that dog don'= t hunt. It doesn't justify multicast either.=0A =0ASo let's improve the dis= cussion here. The Internet, for the forseeable future, at the edge, is unic= ast.=0A =0ACDN's reduce costs for big companies like Netflix, not because t= hey are "close to the watcher" but because it is easier to not have everyth= ing centralized in a single point of failure and needing a huge pipe becaus= e all customers aggregate demand into their central HQ.=0A =0AThere are som= e advantages at the top level peering in distributing CDN sites into the ne= tworks of various access providers (like Comcast) because you can get cheap= er pricing.=0A =0ABut lets think about CDNs in space. They don't deal with = the actual bottleneck in each satellite to ground. The uplink from central = ground stations has LOTS of capacity, it won't be the bottleneck if Starlin= k or any other satellite service for that matter balances its design.=0A = =0AAnd if they get inter-sat links (laser or maybe RF, if the interference = problem with other satellites is resolved at the WRC or other international= body that regulates RF in space) working, where are the CDNs going to main= tain state? The reachable ones for any satellite in the constellation will = be moving relative to the user's dishy, unless it the CDN is in geosync orb= it).=0A =0AStarlink isn't a media company. It doesn't want to own all the c= ontent, or even host all the content.=0A =0ASo I'm pretty darn skeptical!= =0A =0AAt the moment, Starlink is a "last mile" service in a virtual sense.= It connects from the public internet at some pretty high performance point= of presence, hopefully has a pretty good latency, and doesn't let its user= s saturate any satellite (lest huge queueing delay build up). Given that, a= nywhere there is a ground station the public fiber backbone can reach all t= he CDNs.=0A =0AIt's hard to capitalize a last mile fiber service. Each home= passed costs about $2000, if we want to connect up (that includes almost a= ll actual rural residences, but not all the uninhabited parts of the planet= ). For whatever reason, the telcos want no one else to put in fiber, but al= so don't see much profit in extending their coverage, either, because they = aren't allowed to charge for all the content like they used to. Fiber's gre= at advantage is that it has extremely low operational expense.=0A =0AThis i= s what Starlink fills in for. It has lower capex, but HUGE opex. It can't p= erform or scale very well compared to the asymptotic fiber solution. So it = has a pretty good short-term competitive position. And they've gotten very = smart in finding early customers who are eager to buy in.=0A =0AHowever, it= 's important to realize that what drives this all is how the Internet is us= ed and who needs it where, and how much they are willing to pay.=0A =0AOne = thing is clear - Starlink isn't the Internet of the future. It's filling a = niche (a large one, but a niche).=0A =0AThe mistake Motorola made with Irid= ium was in not realizing that cellular telephony was in all respects a bett= er answer, and would be cheaper, too. People did talk about putting CDNs in= space with Iridium, too.=0A =0ABut one needs to understand what CDNs are u= seful for, and really understand what part of the problem that is.=0A =0A ------=_20220831162421000000_81250 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable

Having looked into thi= s a lot, CDNs don't account for very much Internet traffic. There's a lot o= f marketing propaganda out there that suggests this might be true, but when= you try to track down the source, it's almost always from an old slide dec= k presented by a company that sells CDN servers.

=0A

 

=0A

That doesn't mean that CDN servers don't= help a fair bit, but no, today's Internet wouldn't fall apart if we didn't= have CDN servers withint 2 msec. of the edge.

=0A

&= nbsp;

=0A

Also, CDN's need to be BIG to hold all the= videos that people might choose to watch at any particular time.

=0A

 

=0A

So I'm just pointing ou= t that the business case for CDN's in space to merely solve Starlink's pote= ntial issues is probably not great. Maybe 10 years from now. Maybe never. T= he idea that everyone watches TV and the same few seconds of content of a f= ew shows that are extraordinarily popular - well, that dog don't hunt. It d= oesn't justify multicast either.

=0A

 

=0ASo let's improve the discussion here. The Internet, for = the forseeable future, at the edge, is unicast.

=0A

=  

=0A

CDN's reduce costs for big companies like= Netflix, not because they are "close to the watcher" but because it is eas= ier to not have everything centralized in a single point of failure and nee= ding a huge pipe because all customers aggregate demand into their central = HQ.

=0A

 

=0A

There are = some advantages at the top level peering in distributing CDN sites into the= networks of various access providers (like Comcast) because you can get ch= eaper pricing.

=0A

 

=0A

But lets think about CDNs in space. They don't deal with the actual bottle= neck in each satellite to ground. The uplink from central ground stations h= as LOTS of capacity, it won't be the bottleneck if Starlink or any other sa= tellite service for that matter balances its design.

=0A

 

=0A

And if they get inter-sat links (las= er or maybe RF, if the interference problem with other satellites is resolv= ed at the WRC or other international body that regulates RF in space) worki= ng, where are the CDNs going to maintain state? The reachable ones for any = satellite in the constellation will be moving relative to the user's dishy,= unless it the CDN is in geosync orbit).

=0A

 <= /p>=0A

Starlink isn't a media company. It doesn't want = to own all the content, or even host all the content.

=0A

 

=0A

So I'm pretty darn skeptical!

= =0A

 

=0A

At the moment, St= arlink is a "last mile" service in a virtual sense. It connects from the pu= blic internet at some pretty high performance point of presence, hopefully = has a pretty good latency, and doesn't let its users saturate any satellite= (lest huge queueing delay build up). Given that, anywhere there is a groun= d station the public fiber backbone can reach all the CDNs.

=0A

 

=0A

It's hard to capitalize a la= st mile fiber service. Each home passed costs about $2000, if we want to co= nnect up (that includes almost all actual rural residences, but not all the= uninhabited parts of the planet). For whatever reason, the telcos want no = one else to put in fiber, but also don't see much profit in extending their= coverage, either, because they aren't allowed to charge for all the conten= t like they used to. Fiber's great advantage is that it has extremely low o= perational expense.

=0A

 

=0A

This is what Starlink fills in for. It has lower capex, but HUGE opex= . It can't perform or scale very well compared to the asymptotic fiber solu= tion. So it has a pretty good short-term competitive position. And they've = gotten very smart in finding early customers who are eager to buy in.

= =0A

 

=0A

However, it's imp= ortant to realize that what drives this all is how the Internet is used and= who needs it where, and how much they are willing to pay.

=0A

 

=0A

One thing is clear - Starlin= k isn't the Internet of the future. It's filling a niche (a large one, but = a niche).

=0A

 

=0A

The = mistake Motorola made with Iridium was in not realizing that cellular telep= hony was in all respects a better answer, and would be cheaper, too. People= did talk about putting CDNs in space with Iridium, too.

=0A

 

=0A

But one needs to understand what= CDNs are useful for, and really understand what part of the problem that i= s.

=0A

 

=0A

 

<= /font> ------=_20220831162421000000_81250--