Re: [Bloat] cloudflare on a roll

That makes my little ears prick up! --dave On 4/18/23 11:31, Dave Taht via Bloat wrote: https://blog.cloudflare.com/making-home-internet-faster/ -- David Collier-Brown, | Always do right. This will gratify System Programmer and Author | some people and astonish the rest dave.collier-

Re: [Bloat] Hey, all, what about bandlength?

movie footage around the world for processing). To fully utilize their bandwidth, they use protocols that involve lots of connections operating in parallel David Lang On Sun, 9 Apr 2023, Dave Collier-Brown via Bloat wrote: Date: Sun, 9 Apr 2023 16:06:54 -0400 From: Dave Collier-Brown via B

Re: [Bloat] Hey, all, what about bandlength?

too? * Fortunately, no I personally observed 456.2 Mbit/s, about 54% of a gigabit at home, so it's more like the latency cut my bandwidth in half --dave On 4/8/23 22:32, Michael Richardson via Bloat wrote: Dave Collier-Brown <mailto:dave.collier-br...@indexexchange.com> wrote:

Re: [Bloat] Hey, all, what about bandlength?

On 4/8/23 14:49, Michael Richardson wrote: Dave Collier-Brown via Bloat wrote: > They he said "bandlength" > That sounded like an odd name, but the idea was cool: > If I have a bandwidth of 1 Mbit/S, but it takes 2 seconds to deliver 1 > Mbit, do

[Bloat] Hey, all, what about bandlength?

Dave was talking about metrics the other day, attempts like Apple's round-trips per minute. They he said "bandlength" That sounded like an odd name, but the idea was cool: If I have a bandwidth of 1 Mbit/S, but it takes 2 seconds to deliver 1 Mbit, do I have a bandlength of only 1/2 Mbit/S? -

Re: [Bloat] [Starlink] [LibreQoS] Enabling a production model

It can be worse than that: if a monopoly owns the poles, you're going to have to bury your fibre. That will cost you something like $800,000 per mile, more if you have to cross a road. In my home town, Chatham, Ontario, the local ISP is installing fibre underground because the duopoly of cable

Re: [Bloat] Sorry I Missed You! Follow-up for blog.cerowrt.org

The calculator of what speed you need requires me to be in the US, while I'm in Toronto Canada. In principle my work-from-home needs are identical to an employee in New York (:-)) --dave On 3/29/23 09:37, Allconnect Info via Bloat wrote: Hi, Am I sending this to the right contact address at

Re: [Bloat] Offtopic: passive ping. was: Researchers Seeking Probe Volunteers in USA

On 3/13/23 14:50, Sebastian Moeller wrote: [EXTERNAL] This email originated from outside the organization. Do not click links or open attachments unless you recognize the sender and know the content is safe. Hi Dave, On Mar 13, 2023, at 19:22, Dave Collier-Brown via Bloat wrote: On Mon

[Bloat] Offtopic: passive ping. was: Researchers Seeking Probe Volunteers in USA

On Mon, Mar 13, 2023 at 3:02 AM Sebastian Moeller via Starlink wrote: [SM] OK, I will bite, how do you measure achievable throughput without actually generating it? Packet-pair techniques are notoriously imprecise and have funny failure modes. When y

Re: [Bloat] FCC requires broadband "Nutritional Label"

On 11/18/22 07:36, Rich Brown via Bloat wrote: Forwarded from NANOG list: The effective date will be determined later, after publication in the Federal Register and OMB review under PRA. November 17, 2022?The Federal Communications Commission today unveiled new rules that will for the first t

Re: [Bloat] Fwd: Add deadline awareness to QUIC

OK, top-of-head response to the idea of a deadline-aware transport, not the implementation. My company is doing on-line ad auctions, which typically have a deadline of 120 milliseconds, from request to the last packet of the response. The deadline varies on a recipient-by-recipient basis, and

Re: [Bloat] summarizing the bitag latency report?

The paper talks about mechanism, not effect. We need to refactor before we can summarize I've pitched it to my colleagues by saying what problem it causes us, rather than how it does it. Latency pisses Joyce off. When we fetch a page, parts of it appear first, then others, then banner ads, a

[Bloat] Is fair queuing the same thing as fair scheduling?

🔗 https://lwn.net/SubscriberLink/913260/cde661a365fbc233/ John Ousterhout, maybe doing a bit of shit-disturbing??? I think of him as being the tcl/tk guy, not a network nerd... --dave -- David Collier-Brown, | Always do right. T

Re: [Bloat] [Make-wifi-fast] marketing wifi and broadband in the c-suite

Of course, in marketing, one can say "this service responds faster that that one, this one is faster", then pooh-pooh GB/second as, of course, marketing. --dave (:-)) On 8/22/22 15:30, Bob McMahon via Bloat wrote: Thanks for sharing this. It's interesting that the focus is still on speed and n

Re: [Bloat] Researchers discover major roadblock in alleviating network congestion

Yup! My company just decided that routing fallbacks, like the ARPA NIMs had, weren't necessary. Right after the Rogers BGP fiasco that cut off 50% of Canada from the internet... Bother! --dave On 8/4/22 19:46, Daniel Sterling via Bloat wrote: "Flow control power is non-decentralizable" is fro

Re: [Bloat] [Rpm] An 8 years perspective on broadband in the usa - fcc data

It may well* be more diagnostic than downstream latency, it's * commonly the requests that you want to get through, so you ca get downstream flows, and * enthusiastically avoided by the vendors as a metric (;-)) --dave c-b [* polite Canadian for "very much"] On 6/20/22 10:38, Fries, Just

Re: [Bloat] Unconscious Physiological Effects of Search Latency on Users and Their Click Behaviour

Yes, I had a fun argument with a queuing-networks guy over that, with 3 seconds being a red flag in the paper we were looking at. Another was 30 seconds, where a huge plurality of the subjects had forgotten where they were, and had to visibly stop and stare at the screen to figure out what to do

Re: [Bloat] cloudflare consumer labels blog

I added this comment via Hacker News "Packet loss" is more interesting to nerds than consumers, who don't know what it means. Arguably it needs a better name. "Delays due to retransmission" is short enough to fit the label, and is what a packet loss means _to the reader of the label_, rather th

Re: [Bloat] april 7th fcc on broadband labels

A little politicking: change the name to make the metric actually mean what it says. Turn "Typical packet loss" into "Typical retransmissions". "Packet loss" is misleading except to a TCP/IP expert. We recommend reporting reporting what losses cause, retransmissions. Retransmissions are meani

Re: [Bloat] new measurement method

On 2/10/22 10:45, Dave Taht wrote: I'd like a general purpose lagmeter for android and ios. That would be a _far_ better app for the desktop or a phone than a performance (really load) test --dave -- David Collier-Brown, | Always do right. This will gratify System Programmer and Author

Re: [Bloat] [Cerowrt-devel] uplink bufferbloat and scheduling problems

On 12/1/21 16:09, David Lang wrote: On Wed, 1 Dec 2021, David P. Reed wrote: To say it again: More memory *doesn't* improve throughput when the queue depths exceed one packet on average slight disagreement here. the buffer improves throughput up to the point where it handles one burst of pac

Re: [Bloat] Sidebar re illustrating quality (was: New Version Notification for draft-cpaasch-ippm-responsiveness-00.txt)

I like the idea, especially if you did it it as an actual "card", like a school report card On 2021-08-21 6:23 a.m., Erik Auerswald wrote: Perhaps the idea of a "score" (e.g., A through F) assigned to each individual metric that is then combined to produce an "overall score" is not that bad? I

Re: [Bloat] Sidebar re illustrating quality (was: New Version Notification for draft-cpaasch-ippm-responsiveness-00.txt)

Look at the barrel link, in that case: I'll send you a sketch off-list --dave On 2021-08-19 5:17 p.m., Kenneth Porter wrote: --On Thursday, August 19, 2021 2:50 PM -0400 Dave Collier-Brown wrote: I can't actually draw a picture of it here, but there's a good way to show mu

[Bloat] Sidebar re illustrating quality (was: New Version Notification for draft-cpaasch-ippm-responsiveness-00.txt)

On 2021-08-19 11:48 a.m., Christoph Paasch via Bloat wrote: I agree. Just measuring RPM is not useful. As one can have excellent RPM but still have an Internet connection that is barely usable. However, I still believe that a single score for the user would be great (that score would not be RPM

Re: [Bloat] [Rpm] Airbnb

g, 2021, at 10:25 pm, Dave Collier-Brown <mailto:dave.collier-br...@indexexchange.com> wrote: My run of it reported latency, but without any qualifiers... One would reasonable assume that's idle latency, then. - Jonathan Morton _

Re: [Bloat] [Rpm] Airbnb

Does anyone here know M-Labs, which Airbnb uses? They self-describe thusly, "In 2008, Vint Cerf, one of the “fathers of the Internet,” began a series of conversations with Internet researchers... [about] understanding their Internet performance over time. As a result of these conversations, M-

Re: [Bloat] Bechtolschiem

It's written to look like an academic paper, but it's pure marketing.  "Memory is cheap, we used a lot, so let's select some evidence that argues this is a good thing." As always with the coin-operated, the way to get them to change is to offer additional information which * captures their

Re: [Bloat] Terminology for Laypeople

"lag" is often understood by non-technical folks, as in "the lag between the time you step on the gas and the time the car actually speeds up". Some folks who've been exposed to video enough will know about "lag and jitter" (;-)) --dave On 2021-05-12 11:50 a.m., Ingemar Johansson S via Bloat

[Bloat] "Mighty" browser cares about latency

Of course, it's unclear what they've done toward "building a custom low-latency networking protocol", as the blog post at https://blog.mightyapp.com/mightys-secret-plan-to-invent-the-future-of-computing/ is all marketing-announcement razzle-dazzle. --dave David Collier-Brown, 

Re: [Bloat] Questions for Bufferbloat Wikipedia article

To speak to the original question, I'd say bufferbloat * is undesirable latency * was discovered when adding buffers counter-intuitively /slowed /packet flow. That's so as to catch the reader's attention and immediately cast light on the (memorable but mysterious) name. --dave On 2021-

[Bloat] Comparing bufferbloat tests (was: We built a new bufferbloat test and keen for feedback)

On 2020-11-05 6:48 a.m., Toke Høiland-Jørgensen via Bloat wrote: Also, holy cow, what's going on with your connection? The unloaded latency says 17/110/200 min/median/max RTT. Is that due to bad measurements, or do you have a lot of cross traffic and a really bloated link? :/ -Toke The tests

Re: [Bloat] netperf server news

Number? One or two that were complete and successful. Maybe five unsuccessful tries. More of a sequence than a time-period: u{1-5}s{1-2} --daev On 2020-10-06 4:40 p.m., Rich Brown wrote: Thanks for the feedback. Some responses: 1) I'm glad that people are seeing reasonable speeds from the

Re: [Bloat] Adding CAKE "tc qdisc" options to NetworkManager

Alas, once I looked under the covers, it doesn't look easy at all. Definitely not a weekend project. I used to use QEF, and this reminds me of it. --dave On 2020-10-05 1:33 p.m., Jonathan Morton wrote: On 5 Oct, 2020, at 7:13 pm, David Collier-Brown wrote: By pure luck, I ended up chatting w

Re: [Bloat] Adding CAKE "tc qdisc" options to NetworkManager

This appears to be adding a tree of static data structures so nmcli can understand what the options are, including what the defaults are. If so, I can probably stop it after the command parser runs, and not have a huge build-and-test-everything task. --dave On 2020-10-05 1:33 p.m., Jonathan M

[Bloat] An interesting, and probably erroneous, article on 5G and TCP buffering

In "the morning paper", https://blog.acolyer.org/2020/10/05/understanding-operational-5g/, reviewing http://xyzhang.ucsd.edu/papers/DXu_SIGCOMM20_5Gmeasure.pdf With TCP the story is less compelling, with common congestion control algorithms not suiting 5G characteristics Traditional loss/dela

Re: [Bloat] Dumb question time 1: using upload and download speeds from dslreports

Phew, thanks! --dave On 2020-10-04 12:31 p.m., Jonathan Morton wrote: On 4 Oct, 2020, at 6:25 pm, Dave Collier-Brown <mailto:dave.collier-br...@indexexchange.com> wrote: When setting my laptop to explicitly use CAKE for an article- and recipe-writing effort, I blithely took the do

[Bloat] Dumb question time 2: setting qdisc in NetworkManager

Way back in 2018, David P. Reed asked the NetworkManager list about setting qdisc options, but to this day, NW does not document qdisc, and still resets my machine back to fq_codel. I normally don't care, as I'm behind a de-bloated WRT 54, and now a shiny new IQrouter, but I'm writing an artic

[Bloat] Dumb question time 1: using upload and download speeds from dslreports

When setting my laptop to explicitly use CAKE for an article- and recipe-writing effort, I blithely took the download speed and stuffed it into tc qdisc replace dev enp0s25 root cake docsis ack-filter bandwidth 179mbit When Iván Baldo kindly suggested I mention ingress, it suddenly struck me

Re: [Bloat] NeST: Network Stack Tester

Making things easy to test is good, and I generally assume anything said about 5G is marketing (three drink minimum (;-)) Academics often respond positively to opportunities to make their metrics better, so we might want to write to them and suggest some valid follow-on work... to apply for grant

Re: [Bloat] How about a topical LWN article on demonstrating the real-world goodness of CAKE?

Thanks, I may take you up on that, but I still have some blockers to work through with my victims^h^h^h^h^h^h^h colleagues at work. (;-)) --dave On 2020-09-10 11:08 a.m., Anthony Minessale II wrote: Still willing to host a call on SignalWire Work if you want to check it out. I barely use this e

Re: [Bloat] How about a topical LWN article on demonstrating the real-world goodness of CAKE?

Many thanks: will try the skype test later today --dave On 2020-09-07 5:23 a.m., Toke Høiland-Jørgensen wrote: Dave Collier-Brown <mailto:dave.collier-br...@indexexchange.com> writes: LWN said OK, but I'm stuck on the search for a striking test, one that resonates with "gr

Re: [Bloat] How about a topical LWN article on demonstrating the real-world goodness of CAKE?

LWN said OK, but I'm stuck on the search for a striking test, one that resonates with "grandma". My next two thoughts, probably for the current long weekend, is either to call a loop-back number with Skype and/or ask Toke how he got "Big Buck Bunny" to suffer dropouts. I'd love to use the latt

Re: [Bloat] How about a topical LWN article on demonstrating the real-world goodness of CAKE?

Hi Johnathan, I remember you from before! On 2020-08-10 1:58 p.m., Jonathan Foulkes wrote: Because of the degree to which we're working from home and videoconferencing, a lot of low-price, medium-performance devices are suddenly too wimpy for their new role. Big time. As noted above, it not j

Re: [Bloat] [Cerowrt-devel] apparently this is an end goal of a lot of ipv6 work in the ietf

It reads like Dilbert's marketing (three drink minimum) department fired their marcom team and then wrote the presentation (;-)) --dave On 2020-07-02 4:01 p.m., David P. Reed wrote: Interop 2019 gave this an award? I have to say, it reads like a clone of the Bell System Technical Manual (or

Re: [Bloat] What's a good non-intrusive way to look at bloat (and perhaps things like gout (:-))

21 am, Dave Collier-Brown <mailto:dave.collier-br...@indexexchange.com> wrote: We've good tools to measure network performance under stress, by the simple expedient of stressing it, but is there a good approach I could recommend to my company to monitor a bunch of reasonably modern links, w

[Bloat] What's a good non-intrusive way to look at bloat (and perhaps things like gout (:-))

We've good tools to measure network performance under stress, by the simple expedient of stressing it, but is there a good approach I could recommend to my company to monitor a bunch of reasonably modern links, without the measurement significantly affecting their state? I don't mind increasi

Re: [Bloat] kleinrock's work on queue theory is out of print

Worst case, you can borrow them from the Internet archive: https://archive.org/details/queueingsystems01klei https://archive.org/details/queueingsystems00klei --dave On 2020-05-10 1:01 p.m., Dave Taht wrote: I was really shocked to find out today kleinrock's work was out of print. Anyone kno

Re: [Bloat] Fwd: [v3, net-next 1/4] net: qos: introduce a gate control flow action

Does anyone in our community know any of the addressees? --dave On 2020-04-22 9:58 a.m., Dave Taht wrote: to say I don't understand this is an understatement. help? -- Forwarded message - From: Po Liu Date: Tue, Apr 21, 2020 at 8:10 PM Subject: [v3,net-n

[Bloat] Offtopic to: New board that looks interesting

On 2020-04-04 1:36 p.m., Dave Taht wrote: I've been really frustrated with the state of distributions lately, trying to get any complex thing done has required snaps and docker containers and I really prefer running stuff natively when possible. Various language development groups have run int

Re: [Bloat] abc congestion control on time varying wireless links

I like they way they express it. Admittedly, I just explained diminishing returns in advertising in terms of torque and horsepower curves (;-)) --dave On 2019-12-11 2:19 p.m., Prateesh Goyal wrote: Adding Hari, Mohammad On Wed, Dec 11, 2019 at 2:17 PM Dave Taht mailto:dave.t...@gmail.com>> wro

Re: [Bloat] Rigorous Coffee Shop Bloat Testing

Off-topic, my custom phone manufacturer posted a supposed picture of the handset size they expected to use to provide acceptable cooling and battery life: https://upload.wikimedia.org/wikipedia/commons/5/51/BC-611_Radio_Transmitter-Receiver_-_National_Cryptologic_Museum_-_DSC07810.JPG --dave O

Re: [Bloat] [Ecn-sane] My (controversial) position paper on TCP

Super, much appreciated! My thoughts about "effective lifetime" just got a long tail, possibly sufficient to wag the dog (:-)) Do you think we should be shifting to v6, and keeping v4 nearly invariant to address the long tail? --dave On 2019-03-20 6:28 p.m., Jonathan Morton wrote: > I still h

Re: [Bloat] usa bufferbloat

I wonder what Tek Savvy is like? They're a Canadian ISP, based in my old home town, providing their last-mile service over lined controlled by the doupoly (Bell and the local cable monopoly). I know they're smart enough to want to minimize bloat, but I don't know how constrained they are. Anybody

Re: [Bloat] known buffer sizes on switches

On 2018-11-28 11:55 a.m., Dave Taht wrote: > Thank you for that. I do have a long standing dream of a single chip > wifi router, with the lowest SNR possible, and the minimum number of > pins coming off of it. I'd settle for 32MB of (static?) ram on chip as > that has proven sufficient to date to