[Bloat] I am trying to summon the gumption to file on this one

[Dave Taht via Bloat]

Anybody else here care sufficiently?

https://www.broadband.io/c/get-broadband-grant-alerts-news/ntia-issues-request-for-comments-for-the-digital-equity-act-program

-- 
Come Heckle Mar 6-9 at: https://www.understandinglatency.com/
Dave Täht CEO, TekLibre, LLC
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Rpm] [Starlink] [LibreQoS] On FiWi

[David Lang via Bloat]

I'll point out that pre-Internet, there was UUCP and dialups between the 
computers, not even always-on links. So latency was 'wait until the next dialup 
session' and bandwidth was the critical issue.


most of the early applications worked with this environment, so the transition 
to always-connected still didn't have a strong latency driver. It's only as the 
web grew (and other real-time apps were introduced) that latency began to be 
more significant than bulk bandwitdth.


but as you say, people haven't wrapped their heads around 'bandwidth is 
available' yet.


David Lang


On Tue, 21 Mar 2023, rjmcmahon via Bloat wrote:


Date: Tue, 21 Mar 2023 12:58:17 -0700
From: rjmcmahon via Bloat 
Reply-To: rjmcmahon 
To: Frantisek Borsik 
Cc: Dave Taht via Starlink ,
dan , bran...@rd.bbc.co.uk,
libreqos ,
Rpm , bloat 
Subject: Re: [Bloat] [Rpm] [Starlink] [LibreQoS] On FiWi

I was around when BGP & other critical junctures 
https://en.wikipedia.org/wiki/Critical_juncture_theory  the commercial 
internet. Here's a short write-up from another thread with some thoughts 
(Note: there are no queues in the Schramm Model 
https://en.wikipedia.org/wiki/Schramm%27s_model_of_communication )


On why we're here.

I think Stuart's point about not having the correct framing is spot on. 
I also think part of that may come from the internet's origin story 
so-to-speak. In the early days of the commercial internet, ISPs formed 
by buying MODEM banks from suppliers and connecting them to the 
telephone company central offices (thanks Strowger!) and then leasing T1 
lines from the same telco, connecting the two.  Products like a Cisco 
Access Gateway were used for the MODEM side. The 4K independent ISPs 
formed in the U.S. took advantage of statistical multiplexing per IP 
packets to optimize the PSTN's time division multiplexing (TDM) design. 
That design had a lot of extra capacity because of the mother's day 
problem - the network had to carry the peak volume of calls. It was 
always odd to me that the telephone companies basically contracted out 
statistical to TDM coupling of networks and didn't do it themselves. 
This was rectified with broadband and most all the independent ISPs went 
out of business.


IP statistical multiplexing was great except for one thing. The attached 
computers were faster than their network i/o so TCP had to do things 
like congestion control to avoid network collapse based on congestion 
signals (and a very imperfect control loop.) Basically, that extra TDM 
capacity for voice calls was consumed very quickly. This set in motion 
the idea that network channel capacity is a proxy for computer speed as 
when networks are underprovisioned and congested that's basically 
accurate. Van Jacobson's work was most always about congestion on what 
today are bandwidth constrained networks.


This also started a bit of a cultural war colloquially known as 
Bellheads vs Netheads. The human engineers took sides more or less. The 
netheads mostly kept increasing capacity. The market demand curve for 
computer connections drove this. It's come to a head though, in that 
netheads most always overprovisioned similar to solving the mother's day 
problem. (This is different from the electric build out where the goal 
is to drive peak and average loads to merge in order to keep generators 
efficient at a constant speed.)


Many were first stuck with the concept of bandwidth scarcity per those 
origins. But then came bandwidth abundance and many haven't adjusted. 
Mental block number one. Mental block two occurs when one sees all that 
bandwidth and says, let's use it all as it's going to be scarce, like a 
Great Depression-era person hoarding basic items.


A digression; This isn't that much different in the early days before 
Einstein. Einstein changed thinking by realizing that the speed of 
causality was defined or limited by the speed of massless particles, 
i.e. energy or photons. We all come from energy in one way or another. 
So of course it makes sense that our causality system, e.g. aging, is 
determined by that speed. It had to be relative for Maxwell's equations 
to be held true - which Einstein agreed with as true irrelevant of 
inertial frame. A leap for us comes when we realize that the speed of 
causality, i.e. time, is fundamentally the speed of energy.  It's true 
for all clocks, objects, etc. even computers.


So when we engineer systems that queue information, we don't slow down 
energy, we slow down information. Computers are mass information tools 
so slowing down information slows down distributed compute. As Stuart 
says, "It's the latency, stupid".  It's physics too.


I was trying to explain to a dark fiber provider that I wanted 100Gb/s 
SFPs to a residential building in Boston. They said, nobody needs 
100Gb/s and that's correct from a link capacity perspective. But the 
economics & energy required for the lowest latency ber bit delivered 
actually is 100Gb/s SERDES attached t

Re: [Bloat] [Rpm] [Starlink] [LibreQoS] On FiWi

[rjmcmahon via Bloat]

I was around when BGP & other critical junctures 
https://en.wikipedia.org/wiki/Critical_juncture_theory  the commercial 
internet. Here's a short write-up from another thread with some thoughts 
(Note: there are no queues in the Schramm Model 
https://en.wikipedia.org/wiki/Schramm%27s_model_of_communication )


On why we're here.

I think Stuart's point about not having the correct framing is spot on. 
I also think part of that may come from the internet's origin story 
so-to-speak. In the early days of the commercial internet, ISPs formed 
by buying MODEM banks from suppliers and connecting them to the 
telephone company central offices (thanks Strowger!) and then leasing T1 
lines from the same telco, connecting the two.  Products like a Cisco 
Access Gateway were used for the MODEM side. The 4K independent ISPs 
formed in the U.S. took advantage of statistical multiplexing per IP 
packets to optimize the PSTN's time division multiplexing (TDM) design. 
That design had a lot of extra capacity because of the mother's day 
problem - the network had to carry the peak volume of calls. It was 
always odd to me that the telephone companies basically contracted out 
statistical to TDM coupling of networks and didn't do it themselves. 
This was rectified with broadband and most all the independent ISPs went 
out of business.


IP statistical multiplexing was great except for one thing. The attached 
computers were faster than their network i/o so TCP had to do things 
like congestion control to avoid network collapse based on congestion 
signals (and a very imperfect control loop.) Basically, that extra TDM 
capacity for voice calls was consumed very quickly. This set in motion 
the idea that network channel capacity is a proxy for computer speed as 
when networks are underprovisioned and congested that's basically 
accurate. Van Jacobson's work was most always about congestion on what 
today are bandwidth constrained networks.


This also started a bit of a cultural war colloquially known as 
Bellheads vs Netheads. The human engineers took sides more or less. The 
netheads mostly kept increasing capacity. The market demand curve for 
computer connections drove this. It's come to a head though, in that 
netheads most always overprovisioned similar to solving the mother's day 
problem. (This is different from the electric build out where the goal 
is to drive peak and average loads to merge in order to keep generators 
efficient at a constant speed.)


Many were first stuck with the concept of bandwidth scarcity per those 
origins. But then came bandwidth abundance and many haven't adjusted. 
Mental block number one. Mental block two occurs when one sees all that 
bandwidth and says, let's use it all as it's going to be scarce, like a 
Great Depression-era person hoarding basic items.


A digression; This isn't that much different in the early days before 
Einstein. Einstein changed thinking by realizing that the speed of 
causality was defined or limited by the speed of massless particles, 
i.e. energy or photons. We all come from energy in one way or another. 
So of course it makes sense that our causality system, e.g. aging, is 
determined by that speed. It had to be relative for Maxwell's equations 
to be held true - which Einstein agreed with as true irrelevant of 
inertial frame. A leap for us comes when we realize that the speed of 
causality, i.e. time, is fundamentally the speed of energy.  It's true 
for all clocks, objects, etc. even computers.


So when we engineer systems that queue information, we don't slow down 
energy, we slow down information. Computers are mass information tools 
so slowing down information slows down distributed compute. As Stuart 
says, "It's the latency, stupid".  It's physics too.


I was trying to explain to a dark fiber provider that I wanted 100Gb/s 
SFPs to a residential building in Boston. They said, nobody needs 
100Gb/s and that's correct from a link capacity perspective. But the 
economics & energy required for the lowest latency ber bit delivered 
actually is 100Gb/s SERDES attached to lasers attached to fiber.


What we really want is low latency at the lowest energy possible, and 
also to be unleashed from cables (as we're not dogs.) Hence FiWi.


Bob


I do believe that we all want to get the best - latency and speed,
hopefully, in this particular order :-)
The problem was that from the very beginning of the Internet (yeah, I
was still not here, on this planet, when it all started), everything
was optimised for speed, bandwidth and other numbers, but not so much
for bufferbloat in general.
Some of the things that goes into it in the need for speed, are
directly against the fixing latency...and it was not setup for it.
Gamers and Covid (work from home, the need for the enterprise network
but in homes...) brings it into conversation, thankfully, and now we
will deal with it.

Also, there is another thing I see and it's a negative sentiment
against anything busine

Re: [Bloat] [Starlink] Annoyed at 5/1 Mbps...

[Sebastian Moeller via Bloat]

Hi Dan,


> On Mar 21, 2023, at 18:22, dan  wrote:
> 
> GPON is TDMA so the latency is going to be at a minimum the RTT * connected 
> ONUs, vs DSL which is a fixed ratio/scheduler.  

Assuming no proactive grants... are these a thing in PON or only in 
DOCSIS?, but since GPON frames can be shared between ONUs how do you derive the 
"RTT * connected ONUs" formula?


> Standard GPON deployments are typically well over 1 second to the OLT.

I read that as millisecond, which would mean 8 GPON frames... for 
sending the request, processing and arbitrating all requests, assign transmit 
slots and send the transmit maps back to the ONUs, which then actually need to 
send the packets... RTT should not be all that noticeable, at 20 Km the wave 
propagation of light in fiber would be around 2*(2/3 * 3/2)*1000 = 
0.2 milliseconds... (not sure what a realistic maximum length for a PON tree 
is, which probably depends on a number of things anyway, but google says up to 
20 Km for GPON)... but that RTT would be the same for active ethernet...

>  Not that it's bad or anything, but in comparison GPON has very 'wireless' 
> like best case latency but without the wireless variances.

All centrally scheduled link layers will have similar challenges I 
guess?

Kind Regards
Sebastian

> 
> On Tue, Mar 21, 2023 at 6:31 AM Sebastian Moeller  wrote:
> I have to push back gently on this...
> 
> XG(S)-PON is gross 10Gbps (after FEC you are left with around 8,6 Gbps), 
> Noki's proprietary (aka not ITU) @% Gbps PON seems to be abbreviated 25GS-PON.
> 
> Now XGS-PON allows maximally 128 end-nodes in the tree, so:
> 8600/128 = 67.18 Mbps/subscriber
> 
> unless the ISPs royally screwed up the configuration there should be a CIR 
> per subscriber of around 60 Mbps. So setting your cake shaper to 50 Mbps 
> shpuld give you:
> a) 10 times the throughput of the 5/1 Mbps DSL (ignoring overhead 
> compensation for a change, which likely will be in favor of PON)
> b) decent low latency, round robin delay for full MTU packets between 128 
> active nodes would be: 
> packet/sec: ((8.6 * 1000^3)/(1500*8)) = 71.67
> millisec/packet: 1000 / ((8.6 * 1000^3)/(1500*8)) = 0.00139534883721
> round-robin delay 128: 128 * 1000 / ((8.6 * 1000^3)/(1500*8)) = 
> 0.178604651163 milliseconds...
> 
> DSL uses a 4KHz clock so 1000/4000 = 0.25 millisecond quantization
> So XGS-PON has at least theoretical potential to deliver lower latency than 
> DSL, but the details depend on if/how packets are aggregated. HOWEVER the 
> 125µsec GPON frames can be shared between different ONUs in upstream and 
> downstream direction... so these are not a hard quantisation but more the 
> interval between control information required for the access grant cycle...
> 
> c) robustness against RF noise sources and electricity/lightning
> 
> So I am not su sure I would prefer the 5/1 (A)DSL over a PON... 
> 
> That however is orthogonal to me preferring a competent ISP that takes care 
> of keeping latency under load at bay.
> 
> 
> 
> > On Mar 21, 2023, at 12:26, Rich Brown via Starlink 
> >  wrote:
> > 
> > 
> > 
> >> On Mar 21, 2023, at 1:21 AM, Frantisek Borsik via Rpm 
> >>  wrote:
> >> 
> >> Now, I hope to really piss You off with the following statement  :-P but:
> >> 
> >> even sub 5/1 Mbps “broadband” in Africa with bufferbloat fixed on as many 
> >> hops along the internet journey from a data center to the customers mobile 
> >> device (or with just LibreQoS middle box in the ISP’s network) is feeling 
> >> way better than 25Gbps XG-PON. The only time the XG-PON guy could really 
> >> feel like a king of the world would be during his speedtest.
> > 
> > Nope. Sorry - this doesn't piss me off :-) It's just true. 
> > 
> > - 7mbps/768kbps DSL with an IQrouter works fine for two simultaneous Zoom 
> > conferences. (Even though no one would think that it's fast.)
> > - I recommend people on a budget drop their ISP speed so they can afford a 
> > router that does SQM 
> > https://forum.openwrt.org/t/so-you-have-500mbps-1gbps-fiber-and-need-a-router-read-this-first/90305/40
> 
> Even simpler, even on a 100Gbps link nobody stops you from setting 
> your shaper to 50/10 if that is all your router can deliver (and I agree if 
> there are cheaper plans closer to the 50/10 it makes economic sense to scale 
> down the plan)...
> 
> 
> > 
> > The people that get annoyed are those who just upgraded to 1Gbps service 
> > and still are getting fragged in their games.
> > 
> > Rich
> > ___
> > Starlink mailing list
> > starl...@lists.bufferbloat.net
> > https://lists.bufferbloat.net/listinfo/starlink
> 

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Rpm] [Starlink] [LibreQoS] On FiWi

[Frantisek Borsik via Bloat]

I do believe that we all want to get the best - latency and speed,
hopefully, in this particular order :-)
The problem was that from the very beginning of the Internet (yeah, I was
still not here, on this planet, when it all started), everything was
optimised for speed, bandwidth and other numbers, but not so much for
bufferbloat in general.
Some of the things that goes into it in the need for speed, are directly
against the fixing latency...and it was not setup for it. Gamers and Covid
(work from home, the need for the enterprise network but in homes...)
brings it into conversation, thankfully, and now we will deal with it.

Also, there is another thing I see and it's *a negative sentiment against
anything business* (monetisation of, say - lower latency solutions) in
general. If it comes from the general geeky/open source/etc folks, I can
understand it a bit. But it comes also from the business people - assuming
some of You works in big corporations or run ISPs. I'm all against
cronyism, but to throw out the baby with the bathwater - to say that doing
business (i.e. getting paid for delivering something that is missing/fixing
something that is implementing insufficiently) is wrong, to look at it with
disdain, is asinine.

This has the connection with the general "Net Neutrality" (NN) sentiment. I
have 2 suggestions for reading from the other side of the aisle, on this
topic: https://www.martingeddes.com/1261-2
/ (Martin was censored by all major
social media back then, during the days of NN fight in the FCC and
elsewhere.) Second thing is written by one and only Dave Taht:
https://blog.cerowrt.org/post/net_neutrality_customers/

*To conclude, we need to find the way how to benchmark and/or communicate
(translate, if You will) the whole variety of the quality of network
statistics/metrics (which are complex) *like QoE, QoS, latency, jitter,
bufferbloat...to something, that is meaningful for the end user. See this
short proposition of the* Quality of Outcome* by Domos:
https://www.youtube.com/watch?app=desktop&v=MRmcWyIVXvg&t=4185s
There is definitely a lot of work on this - and also on the finding the
right benchmark and its actual measurement side, but it's a step in the
right direction.

*Looking forward to seeing Your take on that proposed Quality of Outcome.
Thanks a lot.*

All the best,

Frank

Frantisek (Frank) Borsik



https://www.linkedin.com/in/frantisekborsik

Signal, Telegram, WhatsApp: +421919416714

iMessage, mobile: +420775230885

Skype: casioa5302ca

frantisek.bor...@gmail.com


On Tue, Mar 21, 2023 at 7:08 PM rjmcmahon via Rpm 
wrote:

> Also, I want my network to be the color clear because I value
> transparency, honesty, and clarity.
>
>
> https://carbuzz.com/news/car-colors-are-more-important-to-buyers-than-you-think
>
> "There are many factors to consider when buying a new car, from price
> and comfort to safety equipment. For many people, color is another
> important factor since it reflects their personality."
>
> "In a study by Automotive Color Preferences 2021 Consumer Survey, 4,000
> people aged 25 to 60 in four of the largest car markets in the world
> (China, Germany, Mexico and the US) were asked about their car color
> preferences. Out of these, 88 percent said that color is a key deciding
> factor when buying a car."
>
> Bob
> > I think we may all be still stuck on numbers. Since infinity is taken,
> > the new marketing number is "infinity & beyond" per Buzz Lightyear
> >
> > Here's what I want, I'm sure others have ideas too:
> >
> > o) We all deserve COPPA. Get the advertiser & their cohorts to stop
> > mining my data & communications - limit or prohibit access to my
> > information by those who continue to violate privacy rights
> > o) An unlimited storage offering with the lowest possible latency paid
> > for annually. That equipment ends up as close as possible to my main
> > home per speed of light limits.
> > o) Security of my network including 24x7x365 monitoring for breaches
> > and for performance
> >  o) Access to any cloud software app. Google & Apple are getting
> > something like 30% for every app on a phone. Seems like a last-mile
> > provider should get a revenue share for hosting apps that aren't being
> > downloaded. Blockbuster did this for DVDs before streaming took over.
> > Revenue shares done properly, while imperfect, can work.
> > o) A life-support capable, future proof, componentized, leash-free,
> > in-home network that is dual-homed over the last mile for redundancy
> > o) Per room FiWi and sensors that can be replaced and upgraded by me
> > ordering and swapping the parts without an ISP getting all my
> > neighbors' consensus & buy in
> > o) VPN capabilities & offerings to the content rights owners'
> > intellectual property for when the peering agreements fall apart
> > o) Video conferencing that works 24x7x365 on all devices
> > o) A single & robust shut-off circuit
> >
> > Bob
> >
> > PS. I think the sweet s

Re: [Bloat] [Starlink] Annoyed at 5/1 Mbps...

[Sebastian Moeller via Bloat]

Hi Jan,

thanks for correcting this. I should have stuck to the point I wanted to make, 
that Nokia's and partner's 25 Gbps PON version is called 25GS-PON.
Even though there is some potential for drama in the Nokia's (simpler encoding 
scheme using?) 25G-PON outside of ITU versus Huawei's (more complex encoding 
schemes using) 50G-PON inside ITU...

Technically I am in no position to decide what a proper standard is and what 
not and whether proprietary versus from a standards body is a useful 
distinction, so I am happy to take your word on it..


> On Mar 21, 2023, at 16:22, Jan Ceuleers via Bloat 
>  wrote:
> 
> On 21/03/2023 13:31, Sebastian Moeller via Bloat wrote:
> (...)
>> Noki's proprietary (aka not ITU) @% Gbps PON seems to be abbreviated 
>> 25GS-PON.
> 
> 25GSPON is in fact not proprietary. It was standardised by means of an
> MSA (multi-source agreement) rather than through the ITU-T, but it is
> standardised and not limited to a single vendor.
> 
> https://www.25gspon-msa.org/
> 
> HTH, Jan
> ___
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Starlink] Annoyed at 5/1 Mbps...

[dan via Bloat]

GPON is TDMA so the latency is going to be at a minimum the RTT * connected
ONUs, vs DSL which is a fixed ratio/scheduler.

Standard GPON deployments are typically well over 1 second to the OLT.  Not
that it's bad or anything, but in comparison GPON has very 'wireless' like
best case latency but without the wireless variances.

On Tue, Mar 21, 2023 at 6:31 AM Sebastian Moeller  wrote:

> I have to push back gently on this...
>
> XG(S)-PON is gross 10Gbps (after FEC you are left with around 8,6 Gbps),
> Noki's proprietary (aka not ITU) @% Gbps PON seems to be abbreviated
> 25GS-PON.
>
> Now XGS-PON allows maximally 128 end-nodes in the tree, so:
> 8600/128 = 67.18 Mbps/subscriber
>
> unless the ISPs royally screwed up the configuration there should be a CIR
> per subscriber of around 60 Mbps. So setting your cake shaper to 50 Mbps
> shpuld give you:
> a) 10 times the throughput of the 5/1 Mbps DSL (ignoring overhead
> compensation for a change, which likely will be in favor of PON)
> b) decent low latency, round robin delay for full MTU packets between 128
> active nodes would be:
> packet/sec: ((8.6 * 1000^3)/(1500*8)) = 71.67
> millisec/packet: 1000 / ((8.6 * 1000^3)/(1500*8)) =
> 0.00139534883721
> round-robin delay 128: 128 * 1000 / ((8.6 * 1000^3)/(1500*8)) =
> 0.178604651163 milliseconds...
>
> DSL uses a 4KHz clock so 1000/4000 = 0.25 millisecond quantization
> So XGS-PON has at least theoretical potential to deliver lower latency
> than DSL, but the details depend on if/how packets are aggregated. HOWEVER
> the 125µsec GPON frames can be shared between different ONUs in upstream
> and downstream direction... so these are not a hard quantisation but more
> the interval between control information required for the access grant
> cycle...
>
> c) robustness against RF noise sources and electricity/lightning
>
> So I am not su sure I would prefer the 5/1 (A)DSL over a PON...
>
> That however is orthogonal to me preferring a competent ISP that takes
> care of keeping latency under load at bay.
>
>
>
> > On Mar 21, 2023, at 12:26, Rich Brown via Starlink <
> starl...@lists.bufferbloat.net> wrote:
> >
> >
> >
> >> On Mar 21, 2023, at 1:21 AM, Frantisek Borsik via Rpm <
> r...@lists.bufferbloat.net> wrote:
> >>
> >> Now, I hope to really piss You off with the following statement  :-P
> but:
> >>
> >> even sub 5/1 Mbps “broadband” in Africa with bufferbloat fixed on as
> many hops along the internet journey from a data center to the customers
> mobile device (or with just LibreQoS middle box in the ISP’s network) is
> feeling way better than 25Gbps XG-PON. The only time the XG-PON guy could
> really feel like a king of the world would be during his speedtest.
> >
> > Nope. Sorry - this doesn't piss me off :-) It's just true.
> >
> > - 7mbps/768kbps DSL with an IQrouter works fine for two simultaneous
> Zoom conferences. (Even though no one would think that it's fast.)
> > - I recommend people on a budget drop their ISP speed so they can afford
> a router that does SQM
> https://forum.openwrt.org/t/so-you-have-500mbps-1gbps-fiber-and-need-a-router-read-this-first/90305/40
>
> Even simpler, even on a 100Gbps link nobody stops you from setting
> your shaper to 50/10 if that is all your router can deliver (and I agree if
> there are cheaper plans closer to the 50/10 it makes economic sense to
> scale down the plan)...
>
>
> >
> > The people that get annoyed are those who just upgraded to 1Gbps service
> and still are getting fragged in their games.
> >
> > Rich
> > ___
> > Starlink mailing list
> > starl...@lists.bufferbloat.net
> > https://lists.bufferbloat.net/listinfo/starlink
>
>
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Starlink] Annoyed at 5/1 Mbps...

[Jan Ceuleers via Bloat]

On 21/03/2023 13:31, Sebastian Moeller via Bloat wrote:
(...)
> Noki's proprietary (aka not ITU) @% Gbps PON seems to be abbreviated 25GS-PON.

25GSPON is in fact not proprietary. It was standardised by means of an
MSA (multi-source agreement) rather than through the ITU-T, but it is
standardised and not limited to a single vendor.

https://www.25gspon-msa.org/

HTH, Jan
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Starlink] Annoyed at 5/1 Mbps...

[Rich Brown via Bloat]

> On Mar 21, 2023, at 8:31 AM, Sebastian Moeller  wrote:
> 
> I have to push back gently on this...
> 
...

> So I am not su sure I would prefer the 5/1 (A)DSL over a PON... 
> 
> That however is orthogonal to me preferring a competent ISP that takes care 
> of keeping latency under load at bay.

OK. I concede. PON (or even a 25/25mbps connection) is way better than DSL. As 
long as I can use a router with SQM :-)

Rich___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Starlink] Annoyed at 5/1 Mbps...

[Sebastian Moeller via Bloat]

I have to push back gently on this...

XG(S)-PON is gross 10Gbps (after FEC you are left with around 8,6 Gbps), Noki's 
proprietary (aka not ITU) @% Gbps PON seems to be abbreviated 25GS-PON.

Now XGS-PON allows maximally 128 end-nodes in the tree, so:
8600/128 = 67.18 Mbps/subscriber

unless the ISPs royally screwed up the configuration there should be a CIR per 
subscriber of around 60 Mbps. So setting your cake shaper to 50 Mbps shpuld 
give you:
a) 10 times the throughput of the 5/1 Mbps DSL (ignoring overhead compensation 
for a change, which likely will be in favor of PON)
b) decent low latency, round robin delay for full MTU packets between 128 
active nodes would be: 
packet/sec: ((8.6 * 1000^3)/(1500*8)) = 71.67
millisec/packet: 1000 / ((8.6 * 1000^3)/(1500*8)) = 0.00139534883721
round-robin delay 128: 128 * 1000 / ((8.6 * 1000^3)/(1500*8)) = 
0.178604651163 milliseconds...

DSL uses a 4KHz clock so 1000/4000 = 0.25 millisecond quantization
So XGS-PON has at least theoretical potential to deliver lower latency than 
DSL, but the details depend on if/how packets are aggregated. HOWEVER the 
125µsec GPON frames can be shared between different ONUs in upstream and 
downstream direction... so these are not a hard quantisation but more the 
interval between control information required for the access grant cycle...

c) robustness against RF noise sources and electricity/lightning

So I am not su sure I would prefer the 5/1 (A)DSL over a PON... 

That however is orthogonal to me preferring a competent ISP that takes care of 
keeping latency under load at bay.



> On Mar 21, 2023, at 12:26, Rich Brown via Starlink 
>  wrote:
> 
> 
> 
>> On Mar 21, 2023, at 1:21 AM, Frantisek Borsik via Rpm 
>>  wrote:
>> 
>> Now, I hope to really piss You off with the following statement  :-P but:
>> 
>> even sub 5/1 Mbps “broadband” in Africa with bufferbloat fixed on as many 
>> hops along the internet journey from a data center to the customers mobile 
>> device (or with just LibreQoS middle box in the ISP’s network) is feeling 
>> way better than 25Gbps XG-PON. The only time the XG-PON guy could really 
>> feel like a king of the world would be during his speedtest.
> 
> Nope. Sorry - this doesn't piss me off :-) It's just true. 
> 
> - 7mbps/768kbps DSL with an IQrouter works fine for two simultaneous Zoom 
> conferences. (Even though no one would think that it's fast.)
> - I recommend people on a budget drop their ISP speed so they can afford a 
> router that does SQM 
> https://forum.openwrt.org/t/so-you-have-500mbps-1gbps-fiber-and-need-a-router-read-this-first/90305/40

Even simpler, even on a 100Gbps link nobody stops you from setting your 
shaper to 50/10 if that is all your router can deliver (and I agree if there 
are cheaper plans closer to the 50/10 it makes economic sense to scale down the 
plan)...


> 
> The people that get annoyed are those who just upgraded to 1Gbps service and 
> still are getting fragged in their games.
> 
> Rich
> ___
> Starlink mailing list
> starl...@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/starlink

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Rpm] [Starlink] [LibreQoS] On FiWi

[Sebastian Moeller via Bloat]

Hi Brandon,


> On Mar 21, 2023, at 01:10, Brandon Butterworth via Rpm 
>  wrote:
> 
> On Mon Mar 20, 2023 at 03:28:57PM -0600, dan via Starlink wrote:
>> I more or less agree with you Frantisek.   There are throughput numbers
>> that are need for current gen and next gen services, but those are often
>> met with 50-100Mbps plans today that are enough to handle multiple 4K
>> streams plus browsing and so forth
> 
> It is for now, question is how busy will it get and will that be before
> the next upgrade round.

I agree these are rates that can work pretty well (assuming the upload 
is wide enough). This is also orthogonal to the point that both copper access 
networks, have already or a close to reaching their reasonable end of life, so 
replacing copper with fiber seems a good idea to future proof the access 
network. But once you do that you realize that actual traffic (at least for big 
ISPs that do not need to buy much transit and get cost neural peerings) is not 
that costly, so offering a 1 Gbps plan instead of a 100 Mbps is a no brainer, 
the customer is unlikely to actually source/sink that much more traffic and you 
might get a few pound/EUR/$ more out of essentially the same load.

> 
> This is why there's a push to sell gigabit in the UK.

I think this also holds for the EU.

> 
> It gives newcomer altnets something the consumers can understand - big
> number - to market against the incumbents sweatng old assets
> with incremental upgrades that will become a problem. From my personal
> point of view (doing active ethernet) it seems pointless making
> equipment more expensive to enable lower speeds to be sold.


One additional reason for the "push for the gigabit" is political in nature. 
The national level of fiber deployment is taken as sort of digital trump game 
in which different countries want to look good, taking available capacity (and 
more so the giga-prefix) as proxy for digitalization and modernity. So if there 
are politic "mandates/desires" to have a high average capacity, then ISPs will 
follow that mandate, especially since that is basically an extension of the 
existing marketing anyways...


>> yet no one talks about latency and packet loss and other useful metrics

Fun fact, I am currently diagnosing issues with my ISP regarding 
packet-loss, one of their gateways produces ~1% packet loss in the download 
direction independent of load, wrecking havoc with speedtest results (Not even 
BBR will tolerate 1% random loss without a noticeable throghuput hit) and hence 
resulting in months of customer complaints the ISP did not manage to root-cause 
and fix... Realistically the packetloss rate without load should be really 
close to 0


> Gamers get it and rate ISPs on it, nobody else cares. Part of the
> reason for throwing bandwith at the home is to ensure the hard to
> replace distribution and house drop is never the problem. Backhaul
> becomes the limit and they can upgrade that more easily when market
> pressure with speedtests show there is a problem.
> 
>> We need a marketing/lobby group.  Not wispa or other individual industry
>> groups, but one specifically for *ISPs that will contribute as well as
>> implement policies and put that out on social media etc etc.  i don't know
>> how we get there without a big player (ie Netflix, hulu..) contributing.
> 
> Peak time congestion through average stream speed reduction is faily obvious
> in playback stats. Any large platform has lots of data on which ISPs
> are performing well.
> 
> We can share stats with the ISPs and tell A that they are performing
> worse than B,C,D if there is a problem. I did want to publish it so
> the public could choose the best but legal were not comfortable
> with that.
> 
> brandon
> ___
> Rpm mailing list
> r...@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/rpm

___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] [Starlink] [Rpm] [LibreQoS] On FiWi

[Brandon Butterworth via Bloat]

On Mon Mar 20, 2023 at 10:21:10PM -0700, Frantisek Borsik wrote:
> Even at Friday evening Netflix time, there?s hardly more than 25/5 Mbps
> consumed.

Today. Today has never been a good target when planning builds that
need to last the next decade. Fibre affords us the luxury of sufficient
capacity to reduce the infrastructure churn where we choose to.

> Also, the real improvements that will be really felt by the people are on
> the bufferbloat front (enterprise as well as residential)

That's a separate matter and needs addressing whatever the delivery
technology and speed.

> If there?s just single one talk that everyone should watch from that
> Understanding Latency webinar series I have shared, it?s this one, with
> Gino Dion (Nokia Bell Labs), Magnus Olden (Domos - Latency Management) and
> Angus Laurie-Pile (GameBench):
> https://m.youtube.com/watch?v=MRmcWyIVXvg&t=1358s
> It?s all about the 1-25Gbps misconception, what we did to put it out there
> as techies, and what can be done to show the customers to change that?40
> minutes, but it?s WORTHWHILE.

TL;DL

I got "how can we monetise latency", says it all, nothing gets fixed
without a premium and the way they were talking that means most do
not get the fix as it becomes an incentive to increase latency to force
more payment. The speed is immaterial in that.

> Now, I hope to really piss You off with the following statement  :-P but:
> 
> even sub 5/1 Mbps ?broadband? in Africa with bufferbloat fixed on as many
> hops along the internet journey from a data center to the customers mobile
> device (or with just LibreQoS middle box in the ISP?s network) is feeling
> way better than 25Gbps XG-PON. The only time the XG-PON guy could really
> feel like a king of the world would be during his speedtest.

So? Some companies will find ways to do things badly regardless, others
make best of what they have. Nothing to get annoyed at nor an argument
to not build faster networks.

I think I may mave missed your point. What are you suggesting, we don't
build faster networks? A new (faster) network build is a great opportunity
to fix bufferbloat.

brandon
___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] Annoyed at 5/1 Mbps...

[Rich Brown via Bloat]

> On Mar 21, 2023, at 1:21 AM, Frantisek Borsik via Rpm 
>  wrote:
> 
> Now, I hope to really piss You off with the following statement  :-P but:
> 
> even sub 5/1 Mbps “broadband” in Africa with bufferbloat fixed on as many 
> hops along the internet journey from a data center to the customers mobile 
> device (or with just LibreQoS middle box in the ISP’s network) is feeling way 
> better than 25Gbps XG-PON. The only time the XG-PON guy could really feel 
> like a king of the world would be during his speedtest.

Nope. Sorry - this doesn't piss me off :-) It's just true. 

- 7mbps/768kbps DSL with an IQrouter works fine for two simultaneous Zoom 
conferences. (Even though no one would think that it's fast.)
- I recommend people on a budget drop their ISP speed so they can afford a 
router that does SQM 
https://forum.openwrt.org/t/so-you-have-500mbps-1gbps-fiber-and-need-a-router-read-this-first/90305/40
 


The people that get annoyed are those who just upgraded to 1Gbps service and 
still are getting fragged in their games.

Rich___
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat