General list for discussing Bufferbloat
 help / color / mirror / Atom feed
From: bob.mcmahon@umbernetworks.com
To: Rich Brown <richb.hanover@gmail.com>
Cc: bloat@lists.bufferbloat.net,
	Igor Aleinikov <igor.aleinikov@adnacom.com>,
	"Dave.seddon Ca" <dave.seddon.ca@gmail.com>,
	Morten <morten@broerup.com>, Jiml <jiml@quicksmart.com>,
	Jason Abele <jason.abele@gmail.com>,
	Shotaro Saito <bobsaito@me.com>,
	Carlos Jones <cjonesii@yahoo.com>, Thomas <thomas@monjalon.net>,
	Robin Jarry <rjarry@redhat.com>,
	Tim Odriscoll <tim.odriscoll@intel.com>,
	Jim <jim@iniholdings.com>, Koen DS <koen0607@gmail.com>
Subject: [Bloat] Re: The hell has frozen: Latency mentioned at APNIC 62, during a keynote!
Date: Fri, 18 Sep 2026 14:41:41 -0700	[thread overview]
Message-ID: <51fcb5e3d6e874c970425c7249980625@umbernetworks.com> (raw)
In-Reply-To: <571D6921-B222-4ECF-B1F9-71184074C5DA@gmail.com>

Hi All,

Just a reminder that bufferbloat isn't the only source of latency or 
jitter. Here's an animation showing some of the other places delay can 
accumulate over the pipeline w/Wi-Fi as a component.

https://www.umbernetworks.com/mac_collapse_lights.php

I'm still working out a few bugs in the animations, but the ideas are 
there. It's from the view of an AP radio or a Fi-Wi radio head, and 
ramps up the traffic from 50 clients.

The animation pipeline includes:

o) Send-side buffering
o) Bytes in flight
o) The 802.11 forwarding plane
o) 802.11 arbitration
o) DMA transfers
o) Over-the-air (OTA) transmission
o) OTA acknowledgment

Bob

PS. Iperf 2 is on the play store.  https://iperf2.fi-wi.com/  Let me 
know if it helps or not.

> We came up with a more accurate, but wordy declaration:
> 
>> "The most important performance metric where
>> systems fall short today is not bandwidth, or compute. It is latency."
> 
> 
> To make a more punchy statement, s/scarce resource/bottleneck/ to 
> get...
> 
>>  "The bottleneck is not bandwidth, or compute. It is latency."
> 
> Thoughts?
> 
> _______________________________________________
> Bloat mailing list -- bloat@lists.bufferbloat.net
> To unsubscribe send an email to bloat-leave@lists.bufferbloat.net

  reply	other threads:[~2026-09-18 21:41 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <178969057349.1938.14005152589460246388@gauss>
2026-09-18 12:11 ` [Bloat] Re: The hell has frozen: Latency mentioned at APNIC 62, during a keynote! Rich Brown
2026-09-18 21:41   ` bob.mcmahon [this message]
2026-09-19 10:03   ` Sebastian Moeller
2026-09-10 17:37 [Bloat] " Frantisek Borsik
2026-09-17 18:24 ` [Bloat] " Stuart Cheshire
2026-09-17 19:56   ` Neal Cardwell
2026-09-17 21:14     ` Richard Scheffenegger
2026-09-18  0:16     ` Stuart Cheshire

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: https://lists.bufferbloat.net/postorius/lists/bloat.lists.bufferbloat.net/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=51fcb5e3d6e874c970425c7249980625@umbernetworks.com \
    --to=bob.mcmahon@umbernetworks.com \
    --cc=bloat@lists.bufferbloat.net \
    --cc=bobsaito@me.com \
    --cc=cjonesii@yahoo.com \
    --cc=dave.seddon.ca@gmail.com \
    --cc=igor.aleinikov@adnacom.com \
    --cc=jason.abele@gmail.com \
    --cc=jim@iniholdings.com \
    --cc=jiml@quicksmart.com \
    --cc=koen0607@gmail.com \
    --cc=morten@broerup.com \
    --cc=richb.hanover@gmail.com \
    --cc=rjarry@redhat.com \
    --cc=thomas@monjalon.net \
    --cc=tim.odriscoll@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox