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

Christoph Paasch cpaasch at apple.com
Thu Jan 5 16:19:20 EST 2023


I think, the cloudflare test uses a single TCP connection, while Ookla/Fast/… use multiple connections.

That’s probably the difference for you.


Christoph

> On Jan 4, 2023, at 2:21 PM, David P. Reed via Starlink <starlink at lists.bufferbloat.net> wrote:
> 
> I don't know how to debug this, but the cloudflare speed test really sucks on my home wired network, compared to others. And also I discovered to my chagrin that the DSLReports speedtest is now broken, at least in my Chrome browser.
>  
> The speeds measured by Cloudflare are essentially 1/2 of both Ookla and Fast.com.
>  
> Cloudflare in my test config gives ~500/10 Mb/sec, with a download latency of 14.6 msec, upload latency of 11.6.
>  
> Oookla and Fast give ~980/25 Mb/sec, with "latency" being about 11 or 12 msec.
>  
> This difference is observed over a tuned "cake" install on my Linux router, and the home network is 10 GigE to my workstation from the router, and the router talks to my DOCSIS cable modem on RCN at 1 GigE, with RCN's product offering being its "Gig" product.
>  
> Now, I hate using Ookla and getting bombarded with ads! I don't really trust fast.com, but it used to be OK.
>  
> The real disappointment was that the DSL Reports speed test, which I used to recommend is so broken. It claims it can't reach any of its 3 selected test sites, because I may have an "alien script" and suggests my DNS might be "slow" or my Chrome might have browser malware.
>  
> Now, maybe Chrome has browser malware on my machine. This is troubling to a serious degree to me, and I will be investigating.
>  
> However, Cloudflare seems to be somewhat flaky to a significant degree. (It also doesn't seem to push a fast network connection nearly hard enough to measure lag under load, it seems to me).
>  
>  
>  
> On Wednesday, January 4, 2023 2:20pm, starlink-request at lists.bufferbloat.net said:
> 
> > Send Starlink mailing list submissions to
> > starlink at lists.bufferbloat.net
> > 
> > To subscribe or unsubscribe via the World Wide Web, visit
> > https://lists.bufferbloat.net/listinfo/starlink
> > or, via email, send a message with subject or body 'help' to
> > starlink-request at lists.bufferbloat.net
> > 
> > You can reach the person managing the list at
> > starlink-owner at lists.bufferbloat.net
> > 
> > When replying, please edit your Subject line so it is more specific
> > than "Re: Contents of Starlink digest..."
> > 
> > 
> > Today's Topics:
> > 
> > 1. Re: [Rpm] the grinch meets cloudflare's christmas present
> > (jf at jonathanfoulkes.com)
> > 
> > 
> > ----------------------------------------------------------------------
> > 
> > Message: 1
> > Date: Wed, 4 Jan 2023 14:20:15 -0500
> > From: "jf at jonathanfoulkes.com" <jf at jonathanfoulkes.com>
> > To: Dave Taht <dave.taht at gmail.com>
> > Cc: bloat <bloat at lists.bufferbloat.net>, libreqos
> > <libreqos at lists.bufferbloat.net>, Cake List
> > <cake at lists.bufferbloat.net>, Dave Taht via Starlink
> > <starlink at lists.bufferbloat.net>, Rpm <rpm at lists.bufferbloat.net>,
> > IETF IPPM WG <ippm at ietf.org>
> > Subject: Re: [Starlink] [Rpm] the grinch meets cloudflare's christmas
> > present
> > Message-ID: <845161E4-474C-44A9-92D4-1702748A3DA1 at jonathanfoulkes.com>
> > Content-Type: text/plain; charset="utf-8"
> > 
> > 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:
> > 
> > -------------- next part --------------
> > A non-text attachment was scrubbed...
> > Name: CFSpeedTest_Gig35_20230104.png
> > Type: image/png
> > Size: 379539 bytes
> > Desc: not available
> > URL:
> > <https://lists.bufferbloat.net/pipermail/starlink/attachments/20230104/d89867fe/attachment.png>
> > -------------- next part --------------
> > 
> > 
> > And an Ookla test run just afterward:
> > 
> > -------------- next part --------------
> > A non-text attachment was scrubbed...
> > Name: Speedtest_net_Gig35_20230104.png
> > Type: image/png
> > Size: 40589 bytes
> > Desc: not available
> > URL:
> > <https://lists.bufferbloat.net/pipermail/starlink/attachments/20230104/d89867fe/attachment-0001.png>
> > -------------- next part --------------
> > 
> > 
> > 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
> > 
> > 
> > ------------------------------
> > 
> > Subject: Digest Footer
> > 
> > _______________________________________________
> > Starlink mailing list
> > Starlink at lists.bufferbloat.net
> > https://lists.bufferbloat.net/listinfo/starlink
> > 
> > 
> > ------------------------------
> > 
> > End of Starlink Digest, Vol 22, Issue 10
> > ****************************************
> >
> _______________________________________________
> Starlink mailing list
> Starlink at lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/starlink

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.bufferbloat.net/pipermail/starlink/attachments/20230105/d549fed5/attachment.html>


More information about the Starlink mailing list