[Bloat] [Cake] [Make-wifi-fast] dslreports is no longer free

Wed May 6 04:19:05 EDT 2020

Hi Sergey,

> On May 4, 2020, at 19:04, Sergey Fedorov <sfedorov at netflix.com> wrote:
> 
> Sergey - I wasn't assuming anything about fast.com. The document you shared wasn't clear about the methodology's details here. Others sadly, have actually used ICMP pings in the way I described. I was making a generic comment of concern.
> 
> That said, it sounds like what you are doing is really helpful (esp. given that your measure is aimed at end user experiential qualities).
> David - my apologies, I incorrectly interpreted your statement as being said in context of fast.com measurements. The blog post linked indeed doesn't provide the latency measurement details - was written before we added the extra metrics. We'll see if we can publish an update. 
> 
> 1) a clear definition of lag under load that is from end-to-end in latency, and involves, ideally, independent traffic from multiple sources through the bottleneck.
> Curious if by multiple sources you mean multiple clients (devices) or multiple connections sending data? 

	Not trying to speak for David obviously, but the dslreports speedtest, when using multiple streams mostly recruited streams for different server locations and reported these locations in some of the detailed report parts. For normal use that level of detail is overkill, but for problematic cases that was really elucidating (the reported the retransmit count for up to 5 server sites):

Server								Nett	Speed Avg	RTT / Jitter Avg	Re-xmit Avg	Cwnd
Singapore (softlayer)				d1	7.3 Mb/s			200.5±7ms		0.1%		154
Houston, USA (softlayer)			d3	3.07 Mb/s		157.6±3.6ms		0.4%		125
Dallas, USA (softlayer)				d3	2.65 Mb/s		150.1±3.3ms		0.6%		131
San Jose, USA (softlayer)			d3	2.77 Mb/s		185.6±5ms		0.5%		126
Nashville, TN, USA (Twinlakes coop)	d3	2.34 Mb/s		127.6±4ms		0.6%		76

Run Log:
0.00s setting download file size to 40mb max for Safari
0.00s Start testing DSL
00.43s Servers available: 10
00.46s pinging 10 locations
01.66s geo location failed
05.47s 19ms Amsterdam, Netherlands, EU
05.47s 63ms Nashville, TN, USA
05.47s 72ms Dallas, USA
05.47s 75ms Houston, USA
05.47s 89ms San Jose, USA
05.47s 96ms Singapore
05.47s could not reach Silver Spring, MD, USA https://t70.dslreports.com
05.47s could not reach Newcastle, Delaware, USA https://t68.dslreports.com
05.47s could not reach Westland, Michigan, USA https://t67.dslreports.com
05.47s could not reach Beaverton, Oregon, USA https://t69.dslreports.com
05.48s 5 seconds measuring idle buffer bloat
10.96s Trial download normal
10.99s Using GET for upload testing
10.99s preference https set to 1
10.99s preference fixrids set to 1
10.99s preference streamsDown set to 16
10.99s preference dnmethod set to websocket
10.99s preference upmethod set to websocket
10.99s preference upduration set to 30
10.99s preference streamsUp set to 16
10.99s preference dnduration set to 30
10.99s preference bloathf set to 1
10.99s preference rids set to [object Object]
10.99s preference compress set to 1
19.11s  stream0 4.71 megabit Amsterdam, Netherlands, EU
19.11s  stream1 2.74 megabit Dallas, USA
19.11s  stream2 4.68 megabit Singapore
19.11s  stream3 2.23 megabit Dallas, USA
19.11s  stream4 3.31 megabit Houston, USA
19.11s  stream5 3.19 megabit Houston, USA
19.11s  stream6 2.83 megabit Amsterdam, Netherlands, EU
19.11s  stream7 1.13 megabit Dallas, USA
19.11s  stream8 2.15 megabit Amsterdam, Netherlands, EU
19.11s  stream9 2.35 megabit San Jose, USA
19.11s  stream10 1.46 megabit Nashville, TN, USA
19.11s  stream11 1.42 megabit Nashville, TN, USA
19.11s  stream12 2.92 megabit Nashville, TN, USA
19.11s  stream13 2.19 megabit Houston, USA
19.11s  stream14 2.16 megabit San Jose, USA
19.11s  stream15 1.2 megabit San Jose, USA
41.26s End of download testing. Starting upload in 2 seconds
43.27s Capping upload streams to 6 because of download result
43.27s starting websocket upload with 16 streams
43.27s minimum upload speed of 0.3 per stream
43.48s sent first packet to t56.dslreports.com
44.08s sent first packet to t59.dslreports.com
44.48s sent first packet to t59.dslreports.com
44.48s sent first packet to t57.dslreports.com
44.68s sent first packet to t56.dslreports.com
44.78s sent first packet to t58.dslreports.com
44.79s got first reply from t56.dslreports.com 221580
44.98s sent first packet to t58.dslreports.com
45.08s sent first packet to t56.dslreports.com
45.14s got first reply from t59.dslreports.com 221580
45.28s sent first packet to t59.dslreports.com
45.53s got first reply from t59.dslreports.com 155106
45.55s got first reply from t57.dslreports.com 70167
45.78s got first reply from t58.dslreports.com 210501
45.85s got first reply from t56.dslreports.com 162492
45.88s sent first packet to t60.dslreports.com
45.88s sent first packet to t71.dslreports.com
46.00s got first reply from t58.dslreports.com 44316
46.08s sent first packet to t71.dslreports.com
46.26s got first reply from t56.dslreports.com 177264
46.28s sent first packet to t71.dslreports.com
46.41s got first reply from t59.dslreports.com 221580
46.58s sent first packet to t58.dslreports.com
46.88s sent first packet to t60.dslreports.com
46.89s got first reply from t60.dslreports.com 99711
47.08s sent first packet to t60.dslreports.com
47.61s got first reply from t58.dslreports.com 221580
47.93s got first reply from t60.dslreports.com 158799
48.09s got first reply from t60.dslreports.com 107097
62.87s Recording upload  21.45
62.87s Timer drops: frames=0 total ms=0
62.87s END TEST
64.88s Total megabytes consumed: 198.8 (down:155 up:43.8)

Not sure how trust-worthy these numbers were, but high retransmit counts correlated with relative low measured goodput... 

I realize that this level of detail is explicitly out of scope for fast.com, but if you collect similar data, exposing it for interested parties following a chain of links would be swell. I am thinking along the lines of  Dougles Adams' "It was on display in the bottom of a locked filing cabinet stuck in a disused lavatory with a sign on the door saying ‘Beware of the Leopard." here ;)

Best Regards
	Sebastian

> 
> SERGEY FEDOROV
> Director of Engineering
> sfedorov at netflix.com
> 121 Albright Way  |  Los Gatos, CA 95032
> 
> 
> 
> 
> On Sun, May 3, 2020 at 8:07 AM David P. Reed <dpreed at deepplum.com> wrote:
> Thanks Sebastian. I do agree that in many cases, reflecting the ICMP off the entry device that has the external IP address for the NAT gets most of the RTT measure, and if there's no queueing built up in the NAT device, that's a reasonable measure. But...
> 
> However, if the router has "taken up the queueing delay" by rate limiting its uplink traffic to slightly less than the capacity (as with Cake and other TC shaping that isn't as good as cake), then there is a queue in the TC layer itself. This is what concerns me as a distortion in the measurement that can fool one into thinking the TC shaper is doing a good job, when in fact, lag under load may be quite high from inside the routed domain (the home).
> 
> As you point out this unmeasured queueing delay can also be a problem with WiFi inside the home. But it isn't limited to that.
> 
> A badly set up shaping/congestion management subsystem inside the NAT can look "very good" in its echo of ICMP packets, but be terrible in response time to trivial HTTP requests from inside, or equally terrible in twitch games and video conferencing.
> 
> So, for example, for tuning settings with "Cake" it is useless.
> 
> To be fair, usually the Access Provider has no control of what is done after the cable is terminated at the home, so as a way to decide if the provider is badly engineering its side, a ping from a server is a reasonable quality measure of the provider. 
> 
> But not a good measure of the user experience, and if the provider provides the NAT box, even if it has a good shaper in it, like Cake or fq_codel, it will just confuse the user and create the opportunity for a "finger pointing" argument where neither side understands what is going on.
> 
> This is why we need 
> 
> 1) a clear definition of lag under load that is from end-to-end in latency, and involves, ideally, independent traffic from multiple sources through the bottleneck.
> 
> 2) ideally, a better way to localize where the queues are building up and present that to users and access providers.  The flent graphs are not interpretable by most non-experts. What we need is a simple visualization of a sketch-map of the path (like traceroute might provide) with queueing delay measures  shown at key points that the user can understand.
> On Saturday, May 2, 2020 4:19pm, "Sebastian Moeller" <moeller0 at gmx.de> said:
> 
>> Hi David,
>> 
>> in principle I agree, a NATed IPv4 ICMP probe will be at best reflected at the NAT
>> router (CPE) (some commercial home gateways do not respond to ICMP echo requests
>> in the name of security theatre). So it is pretty hard to measure the full end to
>> end path in that configuration. I believe that IPv6 should make that
>> easier/simpler in that NAT hopefully will be out of the path (but let's see what
>> ingenuity ISPs will come up with).
>> Then again, traditionally the relevant bottlenecks often are a) the internet
>> access link itself and there the CPE is in a reasonable position as a reflector on
>> the other side of the bottleneck as seen from an internet server, b) the home
>> network between CPE and end-host, often with variable rate wifi, here I agree
>> reflecting echos at the CPE hides part of the issue.
>> 
>> 
>> 
>>> On May 2, 2020, at 19:38, David P. Reed <dpreed at deepplum.com> wrote:
>>> 
>>> I am still a bit worried about properly defining "latency under load" for a
>> NAT routed situation. If the test is based on ICMP Ping packets *from the server*,
>> it will NOT be measuring the full path latency, and if the potential congestion
>> is in the uplink path from the access provider's residential box to the access
>> provider's router/switch, it will NOT measure congestion caused by bufferbloat
>> reliably on either side, since the bufferbloat will be outside the ICMP Ping
>> path.
>> 
>> Puzzled, as i believe it is going to be the residential box that will respond
>> here, or will it be the AFTRs for CG-NAT that reflect the ICMP echo requests?
>> 
>>> 
>>> I realize that a browser based speed test has to be basically run from the
>> "server" end, because browsers are not that good at time measurement on a packet
>> basis. However, there are ways to solve this and avoid the ICMP Ping issue, with a
>> cooperative server.
>>> 
>>> I once built a test that fixed this issue reasonably well. It carefully
>> created a TCP based RTT measurement channel (over HTTP) that made the echo have to
>> traverse the whole end-to-end path, which is the best and only way to accurately
>> define lag under load from the user's perspective. The client end of an unloaded
>> TCP connection can depend on TCP (properly prepared by getting it past slowstart)
>> to generate a single packet response.
>>> 
>>> This "TCP ping" is thus compatible with getting the end-to-end measurement on
>> the server end of a true RTT.
>>> 
>>> It's like tcp-traceroute tool, in that it tricks anyone in the middle boxes
>> into thinking this is a real, serious packet, not an optional low priority
>> packet.
>>> 
>>> The same issue comes up with non-browser-based techniques for measuring true
>> lag-under-load.
>>> 
>>> Now as we move HTTP to QUIC, this actually gets easier to do.
>>> 
>>> One other opportunity I haven't explored, but which is pregnant with
>> potential is the use of WebRTC, which runs over UDP internally. Since JavaScript
>> has direct access to create WebRTC connections (multiple ones), this makes
>> detailed testing in the browser quite reasonable.
>>> 
>>> And the time measurements can resolve well below 100 microseconds, if the JS
>> is based on modern JIT compilation (Chrome, Firefox, Edge all compile to machine
>> code speed if the code is restricted and in a loop). Then again, there is Web
>> Assembly if you want to write C code that runs in the brower fast. WebAssembly is
>> a low level language that compiles to machine code in the browser execution, and
>> still has access to all the browser networking facilities.
>> 
>> Mmmh, according to https://github.com/w3c/hr-time/issues/56 due to spectre
>> side-channel vulnerabilities many browsers seemed to have lowered the timer
>> resolution, but even the ~1ms resolution should be fine for typical RTTs.
>> 
>> Best Regards
>> Sebastian
>> 
>> P.S.: I assume that I simply do not see/understand the full scope of the issue at
>> hand yet.
>> 
>> 
>>> 
>>> On Saturday, May 2, 2020 12:52pm, "Dave Taht" <dave.taht at gmail.com>
>> said:
>>> 
>>>> On Sat, May 2, 2020 at 9:37 AM Benjamin Cronce <bcronce at gmail.com>
>> wrote:
>>>>> 
>>>>>> Fast.com reports my unloaded latency as 4ms, my loaded latency
>> as ~7ms
>>>> 
>>>> I guess one of my questions is that with a switch to BBR netflix is
>>>> going to do pretty well. If fast.com is using bbr, well... that
>>>> excludes much of the current side of the internet.
>>>> 
>>>>> For download, I show 6ms unloaded and 6-7 loaded. But for upload
>> the loaded
>>>> shows as 7-8 and I see it blip upwards of 12ms. But I am no longer using
>> any
>>>> traffic shaping. Any anti-bufferbloat is from my ISP. A graph of the
>> bloat would
>>>> be nice.
>>>> 
>>>> The tests do need to last a fairly long time.
>>>> 
>>>>> On Sat, May 2, 2020 at 9:51 AM Jannie Hanekom
>> <jannie at hanekom.net>
>>>> wrote:
>>>>>> 
>>>>>> Michael Richardson <mcr at sandelman.ca>:
>>>>>>> Does it find/use my nearest Netflix cache?
>>>>>> 
>>>>>> Thankfully, it appears so. The DSLReports bloat test was
>> interesting,
>>>> but
>>>>>> the jitter on the ~240ms base latency from South Africa (and
>> other parts
>>>> of
>>>>>> the world) was significant enough that the figures returned
>> were often
>>>>>> unreliable and largely unusable - at least in my experience.
>>>>>> 
>>>>>> Fast.com reports my unloaded latency as 4ms, my loaded latency
>> as ~7ms
>>>> and
>>>>>> mentions servers located in local cities. I finally have a test
>> I can
>>>> share
>>>>>> with local non-technical people!
>>>>>> 
>>>>>> (Agreed, upload test would be nice, but this is a huge step
>> forward from
>>>>>> what I had access to before.)
>>>>>> 
>>>>>> Jannie Hanekom
>>>>>> 
>>>>>> _______________________________________________
>>>>>> Cake mailing list
>>>>>> Cake at lists.bufferbloat.net
>>>>>> https://lists.bufferbloat.net/listinfo/cake
>>>>> 
>>>>> _______________________________________________
>>>>> Cake mailing list
>>>>> Cake at lists.bufferbloat.net
>>>>> https://lists.bufferbloat.net/listinfo/cake
>>>> 
>>>> 
>>>> 
>>>> --
>>>> Make Music, Not War
>>>> 
>>>> Dave Täht
>>>> CTO, TekLibre, LLC
>>>> http://www.teklibre.com
>>>> Tel: 1-831-435-0729
>>>> _______________________________________________
>>>> Cake mailing list
>>>> Cake at lists.bufferbloat.net
>>>> https://lists.bufferbloat.net/listinfo/cake
>>>> 
>>> _______________________________________________
>>> Cake mailing list
>>> Cake at lists.bufferbloat.net
>>> https://lists.bufferbloat.net/listinfo/cake
>> 
>> 
> 