[Starlink] [Rpm] the grinch meets cloudflare's christmas present

Fri Jan 6 04:43:18 EST 2023

Hi RR,

> On Jan 6, 2023, at 01:01, Dick Roy <dickroy at alum.mit.edu> wrote:
> 
> Hi Sebastian,
>  
> See below …
>  
> -----Original Message-----
> From: Sebastian Moeller [mailto:moeller0 at gmx.de] 
> Sent: Thursday, January 5, 2023 3:26 AM
> To: Dick Roy
> Cc: Dave Collier-Brown; starlink at lists.bufferbloat.net
> Subject: Re: [Starlink] [Rpm] the grinch meets cloudflare's christmas present
>  
> Hi RR,
>  
>  
> > On Jan 5, 2023, at 04:11, Dick Roy via Starlink <starlink at lists.bufferbloat.net> wrote:
> > 
> >  
> >  
> > From: Starlink [mailto:starlink-bounces at lists.bufferbloat.net] On Behalf Of Dave Collier-Brown via Starlink
> > Sent: Wednesday, January 4, 2023 6:48 PM
> > To: starlink at lists.bufferbloat.net
> > Subject: Re: [Starlink] [Rpm] the grinch meets cloudflare's christmas present
> >  
> > I think using "speed" for "the inverse of delay" is pretty normal English, if technically erroneous when speaking nerd or physicist.
> > 
> > [RR] I’ve not heard of that usage before.  The units aren’t commensurate either. 
> > 
> > Using it for volume? Arguably more like fraudulent...
> > 
> > [RR] I don’t think that was Bob’s intent.  I think “load volume” was meant to be a metaphor for “number of bits/bytes” being transported (“by the semi”).  
> > 
> > That said, aren’t users these days educated on “gigs” which they intuitively understand to be Gigabits per second (or Gbps)?  Oddly enough, that is an expression of “data/information/communication rate” in the appropriate units with the nominal technically correct meaning.  
>  
>       [SM] Gigs would have the following confounds if used without a proper definition:
> a) base10 or base2^10?
> b) giga-what? Bit or Byte
> c) Volume or capacity
> d) if capacity, minimal, average, or maximal?
>  
> I note (again, sorry to sound like a broken record) that the national regulatory agency for networks (Bundes-Netzagentur, short BNetzA) in Germany has some detailed instructions about what information ISPs need to supply to their potential customers pre-sale (seehttps://www.bundesnetzagentur.de/SharedDocs/Downloads/DE/Sachgebiete/Telekommunikation/Unternehmen_Institutionen/Anbieterpflichten/Kundenschutz/Transparenzmaßnahmen/Instruction_for_drawing_up_PIS.pdf?__blob=publicationFile&v=1) where the headlines talk correctly about "data transmission rates" but in the text they occasionally fall back to "speed". They also state: "Data transmission rates must be given in megabits per second (Mbit/s)."
> This is both in response to our "speed" discussion, but also one potential way to clarify b) c) and d) above... given that is official this probably also answers a) (base10 otherwise the text would be "Data transmission rates must be given in mebibits per second (Mibit/s).")

> [RR] My reference to “gigs” was to the ads out nowadays from AT&T about becoming Gagillionaires (“Yes, I am Jurgous. … We know!”) that “now have gig speed wireless from AT&T” so they can play all kinds of VR games. 

	[SM2] Ah, not being in the U.S. that campaign has completely evaded my attention.

> J That said, not sure why BNetzA mandates a particular unit for information rates, but that’s their prerogative I guess.

	[SM2] My bigger point was actually that they use the term "data transmission rates" instead of speed... But to your point the product information sheets are designed specifically to make it easy for consumers to compare different offers (for all values of consumer knowledge of networking and SI-prefixes) and requiring a single unit makes comparison simple and gives less leeway to play games. This BNetzA initiative basically removed the previous "up to XX Mbps" promise of ISPs and especially the interpretation that YY with YY << XX is still the contractually fine as "up to" implies not guarantee.

>  Given that the fundamental unit of information is the answer to a YES/NO question (aka a “bit”), it makes sense to measure information in bits (although trits or any other higher order concept could be used as long as the system accounted for fractions thereofJ) (and sets of bits (aka bytes or really octets) because of ancient computer arcitecturesJ).

	[SM2] I agree, but prefer this to be inherent in the term used, and "gigs" did not carry any of that information for me, but again I had not encountered the AT&T campaign...

>  Since we have pretty much settled on the SI second as the accepted unit of time (and multiples thereof e.g. msec, usec, nsec, etc.), it makes sense to measure information flow in bits/sec or some multiples thereof such as Gbps, Mbps, Kbps, etc. and their byte (really octet) versions GBps, MBps, KBps, etc.. Not sure why BNetzA mandates ONLY one of these, but whatever … J

	[SM2] Again I speculate that this is to allow easy comparison even by consumers that are not "fluent" in interpreting SI-prefixes and that might e.g. not know by heart whether Mb is larger or smaller than Gb, let alone why mB is not equal to Mb... this is a transparency measure foremost aimed at all end-customers (business contract do not fall under that regulation as far as I know).

> As for capacity, remember capacity is not something that is measured. It is a fundamental property (an information rate!) of a communication channel which has no other attributes such as minimal, average, or maximal (unless one is talking about time-varying channels and is wanting to characterize the capacity of the channel over time, but that’s another story).

	[SM] On a shared medium (and let's face it head on sooner or later "internet-access" will become a shared medium) the capacity share each user can relay on is rarely static, most often ISPs oversubscribe links and use traffic shapers to limit the maximum capacity share a user can use, but that share is not guaranteed to be available. BNetzA went as far as defining three different data rates with different expected (temporal) availability as well as a (rather complicated) method how end-users can confirm whether ISPs actually deliver the promised rates.

>  As such, comparing volume and capacity is comparing apples and oranges; one is a size of something (e.g. number of megabytes) and the other is a rate (e.g. MBps) so I am not sure what “Volume or capacity” really means.

	[SM] The term Gigs unlike e.g. Mb/s or Mbps (for Megabits per second) does not intuitively define whether time is taken into account, and over here mobile contracts typically come with a "high-speed Volume" after the consumption of which mobile links fall back to truly atrocious rates like 32Kbps, for these kind of contracts ISPs tend to primarily market the Volume (mostly in GB) and only mention the maximal data rates as secondary information.

> I suspect the concept you may be looking for is “achievable rate” rather than “capacity”.

	[SM2] Actually I always try to first calculate the theoretical upper limit of my links (which I then dub the link's "capacity") and I compare the actual measured rates with the theoretical limit... This works well enough for me, since the consumer links I use typically are dominated by a predictable traffic shaper on the ISP side...

>  Achievable rate IS something that is measureable, and varies with load when channels are shared, etc..  Loosely speaking, achievable rate is always less than or equal to the capacity of a channel. 

	[SM2] Yepp, that is why I try to calculate a capacity estimate (and calculate the resulting throughput on different levels).

Regards
	Sebastian

>  
> HNY,
>  
> RR
>  
> --Sebastian
>  
> > 
> > RR
> > 
> > --dave
> > 
> > On 1/4/23 18:54, Bruce Perens via Starlink wrote:
> >> On the other hand, we would like to be comprehensible to normal users, especially when we want them to press their providers to deal with bufferbloat. Differences like speed and rate would go right over their heads.
> >>  
> >> On Wed, Jan 4, 2023 at 1:16 PM Ulrich Speidel via Starlink <starlink at lists.bufferbloat.net> wrote:
> >>> The use of the term "speed" in communications used to be restricted to the speed of light (or whatever propagation speed one happened to be dealing with. Everything else was a "rate". Maybe I'm old-fashioned but I think talking about "speed tests" muddies the waters rather a lot.
> >>>  
> >>> -- 
> >>> **************************************************************** 
> >>> Dr. Ulrich Speidel
> >>> 
> >>> Department of Computer Science 
> >>> 
> >>> Room 303S.594
> >>> Ph: (+64-9)-373-7599 ext. 85282 
> >>> 
> >>> The University of Auckland
> >>> u.speidel at auckland.ac.nz
> >>> http://www.cs.auckland.ac.nz/~ulrich/
> >>> **************************************************************** 
> >>> From: Starlink <starlink-bounces at lists.bufferbloat.net> on behalf of rjmcmahon via Starlink <starlink at lists.bufferbloat.net>
> >>> Sent: Thursday, January 5, 2023 9:02 AM
> >>> To: jf at jonathanfoulkes.com <jf at jonathanfoulkes.com>
> >>> Cc: Cake List <cake at lists.bufferbloat.net>; IETF IPPM WG <ippm at ietf.org>; libreqos <libreqos at lists.bufferbloat.net>; Dave Taht via Starlink <starlink at lists.bufferbloat.net>; Rpm <rpm at lists.bufferbloat.net>; bloat <bloat at lists.bufferbloat.net>
> >>> Subject: Re: [Starlink] [Rpm] the grinch meets cloudflare's christmas present
> >>>  
> >>> Curious to why people keep calling capacity tests speed tests? A semi at 
> >>> 55 mph isn't faster than a porsche at 141 mph because its load volume is 
> >>> larger.
> >>> 
> >>> Bob
> >>> > HNY Dave and all the rest,
> >>> > 
> >>> > Great to see yet another capacity test add latency metrics to the
> >>> > results. This one looks like a good start.
> >>> > 
> >>> > Results from my Windstream DOCSIS 3.1 line (3.1 on download only, up
> >>> > is 3.0) Gigabit down / 35Mbps up provisioning. Using an IQrouter Pro
> >>> > (an i5 x86) with Cake set for 710/31 as this ISP can’t deliver
> >>> > reliable low-latency unless you shave a good bit off the targets. My
> >>> > local loop is pretty congested.
> >>> > 
> >>> > Here’s the latest Cloudflare test:
> >>> > 
> >>> > 
> >>> > 
> >>> > 
> >>> > And an Ookla test run just afterward:
> >>> > 
> >>> > 
> >>> > 
> >>> > 
> >>> > They are definitely both in the ballpark and correspond to other tests
> >>> > run from the router itself or my (wired) MacBook Pro.
> >>> > 
> >>> > Cheers,
> >>> > 
> >>> > Jonathan
> >>> > 
> >>> > 
> >>> >> On Jan 4, 2023, at 12:26 PM, Dave Taht via Rpm 
> >>> >> <rpm at lists.bufferbloat.net> wrote:
> >>> >> 
> >>> >> Please try the new, the shiny, the really wonderful test here:
> >>> >> https://speed.cloudflare.com/
> >>> >> 
> >>> >> I would really appreciate some independent verification of
> >>> >> measurements using this tool. In my brief experiments it appears - as
> >>> >> all the commercial tools to date - to dramatically understate the
> >>> >> bufferbloat, on my LTE, (and my starlink terminal is out being
> >>> >> hacked^H^H^H^H^H^Hworked on, so I can't measure that)
> >>> >> 
> >>> >> My test of their test reports 223ms 5G latency under load , where
> >>> >> flent reports over 2seconds. See comparison attached.
> >>> >> 
> >>> >> My guess is that this otherwise lovely new tool, like too many,
> >>> >> doesn't run for long enough. Admittedly, most web objects (their
> >>> >> target market) are small, and so long as they remain small and not
> >>> >> heavily pipelined this test is a very good start... but I'm pretty
> >>> >> sure cloudflare is used for bigger uploads and downloads than that.
> >>> >> There's no way to change the test to run longer either.
> >>> >> 
> >>> >> I'd love to get some results from other networks (compared as usual to
> >>> >> flent), especially ones with cake on it. I'd love to know if they
> >>> >> measured more minimum rtts that can be obtained with fq_codel or cake,
> >>> >> correctly.
> >>> >> 
> >>> >> Love Always,
> >>> >> The Grinch
> >>> >> 
> >>> >> --
> >>> >> This song goes out to all the folk that thought Stadia would work:
> >>> >> https://www.linkedin.com/posts/dtaht_the-mushroom-song-activity-6981366665607352320-FXtz
> >>> >> Dave Täht CEO, TekLibre, LLC
> >>> >> <image.png><tcp_nup-2023-01-04T090937.211620.LTE.flent.gz>_______________________________________________
> >>> >> Rpm mailing list
> >>> >> Rpm at lists.bufferbloat.net
> >>> >> https://lists.bufferbloat.net/listinfo/rpm
> >>> > 
> >>> > 
> >>> > _______________________________________________
> >>> > Rpm mailing list
> >>> > Rpm at lists.bufferbloat.net
> >>> > https://lists.bufferbloat.net/listinfo/rpm
> >>> _______________________________________________
> >>> Starlink mailing list
> >>> Starlink at lists.bufferbloat.net
> >>> https://lists.bufferbloat.net/listinfo/starlink
> >>> _______________________________________________
> >>> Starlink mailing list
> >>> Starlink at lists.bufferbloat.net
> >>> https://lists.bufferbloat.net/listinfo/starlink
> >> 
> >>  
> >> -- 
> >> Bruce Perens K6BP
> >> 
> >> 
> >> 
> >> _______________________________________________
> >> Starlink mailing list
> >> Starlink at lists.bufferbloat.net
> >> https://lists.bufferbloat.net/listinfo/starlink
> > -- 
> > David Collier-Brown,         | Always do right. This will gratify
> > System Programmer and Author | some people and astonish the rest
> > dave.collier-brown at indexexchange.com |              -- Mark Twain
> >  
> > CONFIDENTIALITY NOTICE AND DISCLAIMER : This telecommunication, including any and all attachments, contains confidential information intended only for the person(s) to whom it is addressed. Any dissemination, distribution, copying or disclosure is strictly prohibited and is not a waiver of confidentiality. If you have received this telecommunication in error, please notify the sender immediately by return electronic mail and delete the message from your inbox and deleted items folders. This telecommunication does not constitute an express or implied agreement to conduct transactions by electronic means, nor does it constitute a contract offer, a contract amendment or an acceptance of a contract offer. Contract terms contained in this telecommunication are subject to legal review and the completion of formal documentation and are not binding until same is confirmed in writing and has been signed by an authorized signatory.
> > 
> > _______________________________________________
> > Starlink mailing list
> > Starlink at lists.bufferbloat.net
> > https://lists.bufferbloat.net/listinfo/starlink