[Starlink] speedtest.net takes a look at sat internet around the globe

[Starlink] speedtest.net takes a look at sat internet around the globe

[Darrell Budic]

[-- Attachment #1: Type: text/plain, Size: 432 bytes --]

https://www.speedtest.net/insights/blog/starlink-hughesnet-viasat-performance-q2-2021/ <https://www.speedtest.net/insights/blog/starlink-hughesnet-viasat-performance-q2-2021/>

Nothing we didn’t know, but interesting comparisons between the 3 sat companies and fixed breadboard around the world.

Made me wonder if there’s anyone else contributing to the speed tests in Iowa county, WI, looks a lot like my averages there...

[-- Attachment #2: Type: text/html, Size: 754 bytes --]

Re: [Starlink] speedtest.net takes a look at sat internet around the globe

[Ulrich Speidel]

[-- Attachment #1: Type: text/plain, Size: 2948 bytes --]

It always pains me to see "speed" tests like these, especially if the 
methodology they've used isn't clear. My big gripes:

- I strongly suspect that the speed tests here (and in so many other 
blogs/vlogs) are UDP-based, which doesn't tell me a thing about how much 
TCP goodput I'll get over any of these links.
- Latency is measured between end user and ... what? The satellite 
gateway? Some imaginary fixed point on the Internet that all our traffic 
has to to through? Or maybe just speedtest.net's servers, whose 
locations possibly don't matter one iota for my Internet performance?
- If we reasonably assume that the capacity of a Starlink satellite 
needs to be shared between its users, then few users / satellite equates 
to a large share of the capacity. From Starlink's front page: "Starlink 
is available to a limited number of users per coverage area at this 
time." Guess what? What we see here may not last, but it's sure great 
for marketing.
- Ever wondered why Starlink's bulk of beta users sits between 
40-something and 50-something degrees of latitude? That's right, because 
that's where you get the largest concentration of satellites right now, 
which helps keep the number of users per satellite down. Elsewhere? 
Tough luck.
- At the other end of your terrestrial broadband connection might be a 
few CDN servers, meaning you and your fellow customers will only need to 
use the ISP's feed once for that viral cat video everyone wants to 
watch. Starlink goes direct to site, not to a local ISP. So if your ISP 
is in space and the CDNs are on the ground, a thousand Starlink users on 
a satellite wanting to watch the cat video will need to bring it across 
the satellite a thousand times. Your terrestrial ISP (or even sat-based 
ISP with a terrestrial network connecting end users) only needs to do 
this once.

But big numbers always look great, don't they?

On 13/08/2021 9:22 am, Darrell Budic wrote:
> https://www.speedtest.net/insights/blog/starlink-hughesnet-viasat-performance-q2-2021/ 
> <https://www.speedtest.net/insights/blog/starlink-hughesnet-viasat-performance-q2-2021/> 
>
>
> Nothing we didn’t know, but interesting comparisons between the 3 sat 
> companies and fixed breadboard around the world.
>
> Made me wonder if there’s anyone else contributing to the speed tests 
> in Iowa county, WI, looks a lot like my averages there...
>
> _______________________________________________
> Starlink mailing list
> Starlink@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/starlink
>
-- 
****************************************************************
Dr. Ulrich Speidel

School of Computer Science

Room 303S.594 (City Campus)
Ph: (+64-9)-373-7599 ext. 85282

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




[-- Attachment #2: Type: text/html, Size: 4121 bytes --]

Re: [Starlink] speedtest.net takes a look at sat internet around the globe

[Ricky Mok]

[-- Attachment #1: Type: text/plain, Size: 4607 bytes --]

I think the blog used the data they collected from speedtest.net. I 
conducted research on that topic. Ookla's speedtest is TCP based 
(usually 4-6 concurrent TCP flows to a test server). Ookla deployed over 
12k test servers (many ISPs contributed to it). The test used IP 
geolocation to pick a set of ten nearest servers geographically and 
establish WebSocket to send "ping"s to test the network latency. The 
test then chooses the server with the lowest latency and send 10 more 
probes to test the RTTs. The "latency" they referred should be the 
extract from the RTT tests.

I think it is a norm to have shared last mile and advertise the "max" 
capacity, just like what cable providers did.

I remembered when they start launching the beta test. Only that latitude 
range had continuous satellite coverage. I assume the service outside 
that range would be unstable. But they have ground station built out of 
that range already. So, they just need to launch more SpaceX rockets...

For CDN, they signed agreements with Google (and Azure?) I expect ground 
stations have direct interconnection with the cloud providers for 
content. They can also have caches installed in the ground station 
(e.g., netflix, youtube), similar to some ordinary ISPs installed 
in-network caches in Central Offices. However, i agree that they cannot 
do caching on the satellites, particularly we have HTTPS :)

Ricky

On 8/12/2021 3:34 PM, Ulrich Speidel wrote:
>
> It always pains me to see "speed" tests like these, especially if the 
> methodology they've used isn't clear. My big gripes:
>
> - I strongly suspect that the speed tests here (and in so many other 
> blogs/vlogs) are UDP-based, which doesn't tell me a thing about how 
> much TCP goodput I'll get over any of these links.
> - Latency is measured between end user and ... what? The satellite 
> gateway? Some imaginary fixed point on the Internet that all our 
> traffic has to to through? Or maybe just speedtest.net's servers, 
> whose locations possibly don't matter one iota for my Internet 
> performance?
> - If we reasonably assume that the capacity of a Starlink satellite 
> needs to be shared between its users, then few users / satellite 
> equates to a large share of the capacity. From Starlink's front page: 
> "Starlink is available to a limited number of users per coverage area 
> at this time." Guess what? What we see here may not last, but it's 
> sure great for marketing.
> - Ever wondered why Starlink's bulk of beta users sits between 
> 40-something and 50-something degrees of latitude? That's right, 
> because that's where you get the largest concentration of satellites 
> right now, which helps keep the number of users per satellite down. 
> Elsewhere? Tough luck.
> - At the other end of your terrestrial broadband connection might be a 
> few CDN servers, meaning you and your fellow customers will only need 
> to use the ISP's feed once for that viral cat video everyone wants to 
> watch. Starlink goes direct to site, not to a local ISP. So if your 
> ISP is in space and the CDNs are on the ground, a thousand Starlink 
> users on a satellite wanting to watch the cat video will need to bring 
> it across the satellite a thousand times. Your terrestrial ISP (or 
> even sat-based ISP with a terrestrial network connecting end users) 
> only needs to do this once.
>
> But big numbers always look great, don't they?
>
> On 13/08/2021 9:22 am, Darrell Budic wrote:
>> https://www.speedtest.net/insights/blog/starlink-hughesnet-viasat-performance-q2-2021/ 
>> <https://www.speedtest.net/insights/blog/starlink-hughesnet-viasat-performance-q2-2021/> 
>>
>>
>> Nothing we didn’t know, but interesting comparisons between the 3 sat 
>> companies and fixed breadboard around the world.
>>
>> Made me wonder if there’s anyone else contributing to the speed tests 
>> in Iowa county, WI, looks a lot like my averages there...
>>
>> _______________________________________________
>> Starlink mailing list
>> Starlink@lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/starlink
>>
> -- 
> ****************************************************************
> Dr. Ulrich Speidel
>
> School of Computer Science
>
> Room 303S.594 (City Campus)
> Ph: (+64-9)-373-7599 ext. 85282
>
> The University of Auckland
> ulrich@cs.auckland.ac.nz  
> http://www.cs.auckland.ac.nz/~ulrich/
> ****************************************************************
>
>
>
>
> _______________________________________________
> Starlink mailing list
> Starlink@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/starlink

[-- Attachment #2: Type: text/html, Size: 6547 bytes --]

Re: [Starlink] speedtest.net takes a look at sat internet around the globe

[David Lang]

[-- Attachment #1: Type: text/plain, Size: 3818 bytes --]

On Fri, 13 Aug 2021, Ulrich Speidel wrote:

> - If we reasonably assume that the capacity of a Starlink satellite needs to 
> be shared between its users, then few users / satellite equates to a large 
> share of the capacity. From Starlink's front page: "Starlink is available to 
> a limited number of users per coverage area at this time." Guess what? What 
> we see here may not last, but it's sure great for marketing.

although as they launch more satellites and implement routing between satellites 
this will also expand

> - Ever wondered why Starlink's bulk of beta users sits between 40-something 
> and 50-something degrees of latitude? That's right, because that's where you 
> get the largest concentration of satellites right now, which helps keep the 
> number of users per satellite down. Elsewhere? Tough luck.

the problem is having satellties overhead for continuous coverage. This happened 
first a higher latitudes. I'm in the Los Angles area and according to a check I 
did a couple weeks ago, there are still 15-75min/day that I would not have any 
satellites available to use. They have launched enough satellites for global 
coverage, but the last several batches are still climbing to their final orbit 
(it takes a couple months post-launch to do this)

> - At the other end of your terrestrial broadband connection might be a few 
> CDN servers, meaning you and your fellow customers will only need to use the 
> ISP's feed once for that viral cat video everyone wants to watch. Starlink 
> goes direct to site, not to a local ISP. So if your ISP is in space and the 
> CDNs are on the ground, a thousand Starlink users on a satellite wanting to 
> watch the cat video will need to bring it across the satellite a thousand 
> times. Your terrestrial ISP (or even sat-based ISP with a terrestrial network 
> connecting end users) only needs to do this once.

bzz, your local ISP needs to transport the data to each customer individually, 
multicast is not a functional thing on the Internet

And with SpaceX putting ground stations on the roofs of major datacenters 
(google and others), they will have as good or better connections than your 
local ISP.

> But big numbers always look great, don't they?

you seem to be wnting to compare starlink to gig fiber to the home or something 
like that.

My sister is the perfect example of their real target. Before I got her 
Starlink, she had the choice between dialup (with toll charges), a wireless ISP 
at symmetrical 2Mb/s, or cellular based service (with a tall mast to hold the 
cell antenna)

starlink isn't targeted at urban areas, but if you look at population density 
maps, the vast majority of the country isn't urban, and while the percentage of 
population isn't as lopsided, that's still a LOT of people

I'm in the Los Angeles suburbs and up until the time that spectrum cable 
upgraded their system a couple years ago, the best I could get (short of 
$thousands to run new wires to my house) was 10M/2M dsl. I'm going to get 
starlink as a backup, and as a portable setup that I can take when traveling 
(camping, search and rescue bases, etc)

David Lang

> On 13/08/2021 9:22 am, Darrell Budic wrote:
>> https://www.speedtest.net/insights/blog/starlink-hughesnet-viasat-performance-q2-2021/ 
>> <https://www.speedtest.net/insights/blog/starlink-hughesnet-viasat-performance-q2-2021/> 
>> 
>> Nothing we didn’t know, but interesting comparisons between the 3 sat 
>> companies and fixed breadboard around the world.
>> 
>> Made me wonder if there’s anyone else contributing to the speed tests in 
>> Iowa county, WI, looks a lot like my averages there...
>> 
>> _______________________________________________
>> 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

Re: [Starlink] speedtest.net takes a look at sat internet around the globe

[Ulrich Speidel]

On 13/08/2021 11:24 am, Ricky Mok wrote:
>
> I think the blog used the data they collected from speedtest.net. I 
> conducted research on that topic. Ookla's speedtest is TCP based 
> (usually 4-6 concurrent TCP flows to a test server). Ookla deployed 
> over 12k test servers (many ISPs contributed to it). The test used IP 
> geolocation to pick a set of ten nearest servers geographically and 
> establish WebSocket to send "ping"s to test the network latency. The 
> test then chooses the server with the lowest latency and send 10 more 
> probes to test the RTTs. The "latency" they referred should be the 
> extract from the RTT tests.
>
That's lumping in a lot of other factors here that have little to do 
with the technology: Even if you do TCP-based tests, it's a matter of 
how long you allow these to ramp up for (cwnd opening speed depends 
crucially on RTT between endpoints). How much congestion exists on the 
path between server and last mile - pings don't necessarily tell you 
that. It's also relying on the topological accuracy of geolocation. 
Geographical proximity != topological proximity in the age of BGP and 
tiered ISPs (we've seen a great deal of traffic between NZ and Asia go 
via the mainland US - and this went on for years!). But most 
importantly, you're comparing mature networks with decades of 
underinvestment and a heavy user load against a new network with light 
beta user load.
>
> I think it is a norm to have shared last mile and advertise the "max" 
> capacity, just like what cable providers did.
>
In the case of Starlink, that last mile is awfully long, though...
>
> I remembered when they start launching the beta test. Only that 
> latitude range had continuous satellite coverage. I assume the service 
> outside that range would be unstable. But they have ground station 
> built out of that range already. So, they just need to launch more 
> SpaceX rockets...
>
Indeed. But there is more to this than that. Basically, each satellite 
in an inclined orbit crosses each parallel (latitude) up to its 
inclination twice per orbit. Since there is a lot less length of 
parallel at higher-numbered latitudes, parallels with latitudes close 
the the inclination get more satellite crossings per mile of parallel 
per hour. So more rockets isn't going to change the density disparity 
unless their satellites will go into orbits with vastly different 
inclinations (which I hope they will).
>
> For CDN, they signed agreements with Google (and Azure?) I expect 
> ground stations have direct interconnection with the cloud providers 
> for content. They can also have caches installed in the ground station 
> (e.g., netflix, youtube), similar to some ordinary ISPs installed 
> in-network caches in Central Offices. However, i agree that they 
> cannot do caching on the satellites, particularly we have HTTPS :)
>
That still puts them on the wrong side of the link ;-)

Honestly, I can understand Americans' and Canadian frustrations with 
their ISPs in rural areas, but compared to people who simply cannot get 
Internet from any other source than satellite for reasons of remoteness, 
depth of ocean or impenetrability of jungle or mountains, that 
regulatory problem is very much a first world one ;-)


-- 
****************************************************************
Dr. Ulrich Speidel

School of Computer Science

Room 303S.594 (City Campus)
Ph: (+64-9)-373-7599 ext. 85282

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

Re: [Starlink] speedtest.net takes a look at sat internet around the globe

[David Lang]

On Fri, 13 Aug 2021, Ulrich Speidel wrote:

> Indeed. But there is more to this than that. Basically, each satellite 
> in an inclined orbit crosses each parallel (latitude) up to its 
> inclination twice per orbit. Since there is a lot less length of 
> parallel at higher-numbered latitudes, parallels with latitudes close 
> the the inclination get more satellite crossings per mile of parallel 
> per hour. So more rockets isn't going to change the density disparity 
> unless their satellites will go into orbits with vastly different 
> inclinations (which I hope they will).

the first shell is all the same inclination, but they have launched enough to 
get full coverage even down to the equator. later shells are going to be a 
different inclination.

see the wikipedia page on starlink for the description of the different shells
https://en.wikipedia.org/wiki/Starlink#Constellation_design_and_status. phase 1 
is ~4200 satellites (they've launched ~1700), phase 2 is an additional ~7500 at 
a lower altitude, and they've talked about wanting a phase 3 that would take the 
total count of over 40,000 satellites.

David Lang

Re: [Starlink] speedtest.net takes a look at sat internet around the globe

[Ulrich Speidel]

[-- Attachment #1: Type: text/plain, Size: 2487 bytes --]

>
> > Indeed. But there is more to this than that. Basically, each satellite
> > in an inclined orbit crosses each parallel (latitude) up to its
> > inclination twice per orbit. Since there is a lot less length of
> > parallel at higher-numbered latitudes, parallels with latitudes close
> > the the inclination get more satellite crossings per mile of parallel
> > per hour. So more rockets isn't going to change the density disparity
> > unless their satellites will go into orbits with vastly different
> > inclinations (which I hope they will).
>
> the first shell is all the same inclination, but they have launched 
> enough to
> get full coverage even down to the equator. later shells are going to 
> be a
> different inclination.

I'm quite aware of this, but Starlink have changed their orbital 
configuration repeatedly over the life of the project thus far. But 
having a satellite within range is only one half of the connection. That 
satellite also needs to be able to link you to something.

And so there won't be full coverage around the tropics for a while as 
most Starlink birds at this point in time aren't equipped with laser 
interconnects. They currently work bent pipe only, which means being 
within a few hundred km of a teleport or no connectivity at all. Many 
remote Pacific islands are further than that away from the nearest other 
piece of land, let alone a Starlink teleport.

But yeah, the plans are noted. I'd add to that that longish zig-zagging 
laser forwarding paths pose the question of latency, too, so it'll be 
interesting how that will play out.

>
> see the wikipedia page on starlink for the description of the 
> different shells
> https://en.wikipedia.org/wiki/Starlink#Constellation_design_and_status 
> <https://en.wikipedia.org/wiki/Starlink#Constellation_design_and_status>. 
> phase 1
> is ~4200 satellites (they've launched ~1700), phase 2 is an additional 
> ~7500 at
> a lower altitude, and they've talked about wanting a phase 3 that 
> would take the
> total count of over 40,000 satellites.

Taken as read a long time ago ;-)

-- 

****************************************************************
Dr. Ulrich Speidel

School of Computer Science

Room 303S.594 (City Campus)
Ph: (+64-9)-373-7599 ext. 85282

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




[-- Attachment #2: Type: text/html, Size: 3461 bytes --]

Re: [Starlink] speedtest.net takes a look at sat internet around the globe

[David Lang]

On Fri, 13 Aug 2021, Ulrich Speidel wrote:

> But yeah, the plans are noted. I'd add to that that longish zig-zagging laser 
> forwarding paths pose the question of latency, too, so it'll be interesting 
> how that will play out.

true, but the speed of light in a vaccum is ~30% higher than the speed of light 
in fiber, so it's going to depend on how many hops and the time cost of each 
hop.

David Lang

Re: [Starlink] speedtest.net takes a look at sat internet around the globe

[Ulrich Speidel]

On 13/08/2021 1:54 pm, David Lang wrote:
> On Fri, 13 Aug 2021, Ulrich Speidel wrote:
>
> > - If we reasonably assume that the capacity of a Starlink satellite 
> needs to
> > be shared between its users, then few users / satellite equates to a 
> large
> > share of the capacity. From Starlink's front page: "Starlink is 
> available to
> > a limited number of users per coverage area at this time." Guess 
> what? What
> > we see here may not last, but it's sure great for marketing.
>
> although as they launch more satellites and implement routing between 
> satellites
> this will also expand
I would hope so, but that's a while off yet.
>
> > - Ever wondered why Starlink's bulk of beta users sits between 
> 40-something
> > and 50-something degrees of latitude? That's right, because that's 
> where you
> > get the largest concentration of satellites right now, which helps 
> keep the
> > number of users per satellite down. Elsewhere? Tough luck.
>
> the problem is having satellties overhead for continuous coverage. 
> This happened
> first a higher latitudes. I'm in the Los Angles area and according to 
> a check I
> did a couple weeks ago, there are still 15-75min/day that I would not 
> have any
> satellites available to use. 
Indeed, but for most people, it's not just that you need to be able to 
see a satellite. You need to be in range of a satellite that can also 
see a gateway, and the probability of your Dishy finding one that is 
meeting this criterion goes up with the number of satellites you can see 
and with your proximity to a gateway. As you do have two gateways in the 
Los Angeles area, this condition will be met for most satellites you can 
see, but with often hundreds of miles between gateways even in the 
continental US, the unavailability periods can be significantly larger 
for others at your latitude.
> They have launched enough satellites for global
> coverage, but the last several batches are still climbing to their 
> final orbit
> (it takes a couple months post-launch to do this)
As for overhead coverage, yes, as for connectivity, no - for lack of 
interconnects in the vast majority of the present fleet.
>
>
> bzz, your local ISP needs to transport the data to each customer 
> individually,
> multicast is not a functional thing on the Internet

Nobody is talking about multicasting here! I'm talking about content 
delivery networks, and they're very much a functional thing, and when 
they break, people will claim that "the Internet is down", as happened a 
few times over the last few months.

Much of the conventional satellite Internet into remote places (and that 
can mean entire countries) used to be backhaul to a local ISP, who would 
then distribute via WiFi, DSL, or whatever. When such ISPs are able to 
host CDN servers in their remote location data centre, then the cat 
video served by the CDN will travel across the sat link once and will be 
cached at the local CDN for multiple downloads by local users.

Direct-to-site operators such as Starlink (and yes I realise Hughes and 
Viasat do this, too, albeit via GEO) have to deliver the cat video to 
each end user through the bottleneck resource satellite.
> And with SpaceX putting ground stations on the roofs of major datacenters
> (google and others), they will have as good or better connections than 
> your
> local ISP.
>
> > But big numbers always look great, don't they?
>
> you seem to be wnting to compare starlink to gig fiber to the home or 
> something
> like that.

Well, I'm comparing it to two quite disparate things, really, so let's 
declare:

1) Gig fibre to the home indeed (which you can get here in Auckland, NZ, 
at least in most suburbs, see 
https://comcom.govt.nz/__data/assets/pdf_file/0027/252288/MBNZ-Summer-Report-2021-13-April-2021.pdf), 
connected to the rest of the world by shared submarine fibre capacity 
roughly equivalent to Starlink's entire current global system capacity 
(including satellites over remote regions far away from gateways). 
That's a couple of dozen Tb/s and us five million plus our 20 million 
Aussie mates across the ditch like have required regular capacity 
upgrades in recent years, suggesting that we manage to work it pretty 
well. I'll see whether I can provide some speedtest.net data FYI in a 
later post.

2) The 4000 ms all up RTT GEO satellite connection that my incoming PhD 
student in Tarawa, Kiribati, is on. He'd love your sister's 
connectivity. There are a lot of islands like this in the Pacific, and 
conditions differ. Many use trunk-to-ISP satellite services, sometimes 
to competing ISPs, sometimes to a monopoly ISP. Some use GEO, some use 
O3b MEO. Some now use direct-to-site Kacific HTS GEO. Some may be within 
reach of Starlink, many won't be, and in some cases, the local regulator 
forbids competition to the monopoly telco and won't allow direct-to-site 
services. And in some places there's hope for change and in others there 
are lock-in contracts running for years into the future.

>
> My sister is the perfect example of their real target. Before I got her
> Starlink, she had the choice between dialup (with toll charges), a 
> wireless ISP
> at symmetrical 2Mb/s, or cellular based service (with a tall mast to 
> hold the
> cell antenna)
>
> starlink isn't targeted at urban areas, but if you look at population 
> density
> maps, the vast majority of the country isn't urban, and while the 
> percentage of
> population isn't as lopsided, that's still a LOT of people
Quite aware of this. There's a nice write-up here of the infrastructural 
problems you face in the US and how Starlink is really designed for that 
market 
https://www.theverge.com/22435030/starlink-satellite-internet-spacex-review 
<https://www.theverge.com/22435030/starlink-satellite-internet-spacex-review> 

>
> I'm in the Los Angeles suburbs and up until the time that spectrum cable
> upgraded their system a couple years ago, the best I could get (short of
> $thousands to run new wires to my house) was 10M/2M dsl. I'm going to get
> starlink as a backup, and as a portable setup that I can take when 
> traveling
> (camping, search and rescue bases, etc)
I was on DSL until about 2-3 years ago. The then government here decided 
that connectivity pretty much everywhere was a strategic investment 
must-have. They'd already split the old monopoly post office telco into 
a network and a retail arm, and invited competition. They then compelled 
(and subsidised) the network company to roll out fibre to wherever this 
was possible - for free - and make it available to ISPs for a fee set by 
the regulator. So they literally came door knocking to connect us up, at 
zero cost (except time to make sure "a foot deep" meant more than 
brushing the leaves aside). At the end of 2022, 87% of New Zealanders 
will have fibre access, two years ahead of target.

-- 
****************************************************************
Dr. Ulrich Speidel

School of Computer Science

Room 303S.594 (City Campus)
Ph: (+64-9)-373-7599 ext. 85282

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

Re: [Starlink] speedtest.net takes a look at sat internet around the globe

[Darrell Budic]

[-- Attachment #1: Type: text/plain, Size: 5922 bytes --]

This matches what I’ve seen from my side. I host a speedtest.net server at the $dayjob in Chicago, with ipv6. That network is peered with Google's, so all of it’s 10G connection is available to it. Of course, it’s way down the list in speedtest servers for me because of the geographic nature of speedtest.net’s recommendations. Those recommended servers are even more useless with a service like starlink, but that doesn’t seem to bother speedtest. 

With v6 over starlink, it does make it a reasonable test of throughput for me, since I have a better than usual understanding of the path and available bandwidth in my test case. Also why I suspect I’m most of the data for Iowa County, WI, I’ve run a fair number of speedtests even without automation :)

  -Darrell

> On Aug 12, 2021, at 6:24 PM, Ricky Mok <cskpmok@caida.org> wrote:
> 
> I think the blog used the data they collected from speedtest.net. I conducted research on that topic. Ookla's speedtest is TCP based (usually 4-6 concurrent TCP flows to a test server). Ookla deployed over 12k test servers (many ISPs contributed to it). The test used IP geolocation to pick a set of ten nearest servers geographically and establish WebSocket to send "ping"s to test the network latency. The test then chooses the server with the lowest latency and send 10 more probes to test the RTTs. The "latency" they referred should be the extract from the RTT tests.
> 
> I think it is a norm to have shared last mile and advertise the "max" capacity, just like what cable providers did.
> 
> I remembered when they start launching the beta test. Only that latitude range had continuous satellite coverage. I assume the service outside that range would be unstable. But they have ground station built out of that range already. So, they just need to launch more SpaceX rockets...
> 
> For CDN, they signed agreements with Google (and Azure?) I expect ground stations have direct interconnection with the cloud providers for content. They can also have caches installed in the ground station (e.g., netflix, youtube), similar to some ordinary ISPs installed in-network caches in Central Offices. However, i agree that they cannot do caching on the satellites, particularly we have HTTPS :)
> 
> Ricky
> 
> On 8/12/2021 3:34 PM, Ulrich Speidel wrote:
>> It always pains me to see "speed" tests like these, especially if the methodology they've used isn't clear. My big gripes:
>> 
>> - I strongly suspect that the speed tests here (and in so many other blogs/vlogs) are UDP-based, which doesn't tell me a thing about how much TCP goodput I'll get over any of these links.
>> - Latency is measured between end user and ... what? The satellite gateway? Some imaginary fixed point on the Internet that all our traffic has to to through? Or maybe just speedtest.net's servers, whose locations possibly don't matter one iota for my Internet performance?
>> - If we reasonably assume that the capacity of a Starlink satellite needs to be shared between its users, then few users / satellite equates to a large share of the capacity. From Starlink's front page: "Starlink is available to a limited number of users per coverage area at this time." Guess what? What we see here may not last, but it's sure great for marketing.
>> - Ever wondered why Starlink's bulk of beta users sits between 40-something and 50-something degrees of latitude? That's right, because that's where you get the largest concentration of satellites right now, which helps keep the number of users per satellite down. Elsewhere? Tough luck.
>> - At the other end of your terrestrial broadband connection might be a few CDN servers, meaning you and your fellow customers will only need to use the ISP's feed once for that viral cat video everyone wants to watch. Starlink goes direct to site, not to a local ISP. So if your ISP is in space and the CDNs are on the ground, a thousand Starlink users on a satellite wanting to watch the cat video will need to bring it across the satellite a thousand times. Your terrestrial ISP (or even sat-based ISP with a terrestrial network connecting end users) only needs to do this once.
>> 
>> But big numbers always look great, don't they?
>> 
>> On 13/08/2021 9:22 am, Darrell Budic wrote:
>>> https://www.speedtest.net/insights/blog/starlink-hughesnet-viasat-performance-q2-2021/ <https://www.speedtest.net/insights/blog/starlink-hughesnet-viasat-performance-q2-2021/>
>>> 
>>> Nothing we didn’t know, but interesting comparisons between the 3 sat companies and fixed breadboard around the world.
>>> 
>>> Made me wonder if there’s anyone else contributing to the speed tests in Iowa county, WI, looks a lot like my averages there...
>>> 
>>> 
>>> _______________________________________________
>>> Starlink mailing list
>>> Starlink@lists.bufferbloat.net <mailto:Starlink@lists.bufferbloat.net>
>>> https://lists.bufferbloat.net/listinfo/starlink <https://lists.bufferbloat.net/listinfo/starlink>
>>> 
>> -- 
>> ****************************************************************
>> Dr. Ulrich Speidel
>> 
>> School of Computer Science
>> 
>> Room 303S.594 (City Campus)
>> Ph: (+64-9)-373-7599 ext. 85282
>> 
>> The University of Auckland
>> ulrich@cs.auckland.ac.nz <mailto:ulrich@cs.auckland.ac.nz> 
>> http://www.cs.auckland.ac.nz/~ulrich/ <http://www.cs.auckland.ac.nz/~ulrich/>
>> ****************************************************************
>> 
>> 
>> 
>> 
>> 
>> _______________________________________________
>> Starlink mailing list
>> Starlink@lists.bufferbloat.net <mailto:Starlink@lists.bufferbloat.net>
>> https://lists.bufferbloat.net/listinfo/starlink <https://lists.bufferbloat.net/listinfo/starlink>
> _______________________________________________
> Starlink mailing list
> Starlink@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/starlink


[-- Attachment #2: Type: text/html, Size: 8571 bytes --]