General list for discussing Bufferbloat
 help / color / mirror / Atom feed
From: Dave Taht <dave.taht@gmail.com>
To: bloat <bloat@lists.bufferbloat.net>
Subject: [Bloat] In terms of bufferbloat-aware buyer constructing a RFP
Date: Tue, 18 May 2021 06:48:46 -0700	[thread overview]
Message-ID: <CAA93jw4Oi2s=yq89aeFJ+PpzAcZYZ_ivj5pdaqqTCvSMJN3x1Q@mail.gmail.com> (raw)

No cloudy provider (I know of) has a SLA regarding bufferbloat related
issues. They sell things by the cpu, and do not
guarantee bandwidth or latencies. No chipmaker tests comprehensibly
for bufferbloat. Most downstream vendors have no clue, nor the users.

Perhaps this document, presented as part of an RFP (request for
proposal) by buyers, would be a start, much like iso 9001 compliance
once was. Or, reconstructed as a survey of the industry, distributed
widely?

https://docs.google.com/document/d/1i9-9M8589pEX_u7vKuwDFoQqbhJRlNPdLqXcxuDP73s/edit?usp=sharing

It's presently short enough to repaste here, but I would prefer
comments and additions be made directly on the url above.

VENDOR BUFFERBLOAT[1] STRATEGY QUESTIONS


What is your bufferbloat-fighting strategy across your product lines?

On Cable?

On DSL?

On Fiber ONTs?

On Ethernet?

On RocE?

On WiFi?

On Powerline Ethernet?

On Thread?

To what extent have you adopted the recommendations of IETF RFC7567 (
https://datatracker.ietf.org/doc/html/rfc7567)

Across your product lines?

In your cloud?

In your end user products?

Are you currently measuring Rate Vs Range Vs TCP RTT Latency across
your wireless and wifi products as google does?

What do you use to benchmark TCP RTT latency under load?

Do you presently have in your offloaded forwarding engines any support
for the following AQM or FQ or FQ+AQM technologies?

RED

HTB

SFQ (dscp)

SFQ (5 tuple)

DRR (dscp)

DRR (5 tuple)

Pie (RFC8033)

DOCSIS-Pie (RFC8034)

Codel (RFC8289)

Fq_codel (RFC8290)

Cake

Fq_pie

Dualpi


Assuming you support one or more of these, do any have RFC3168 ECN
support? DCTCP ECN support?



Have you successfully managed to offload “SQM” (Smart Queue
Management) support in anything?


Can you demonstrate BQL in operation on your Linux based products
during your benchmarks? (tool: bqlmon)


What Linux network scheduler is commonly used in your cloud, or end
user products, or testing regimes?

Sch_pfifo

Sch_fq

Sch_fq_codel

Sch_pie

Sch_tbf

Sch_htb

Sch_cake

Other


What operating systems are used to test with?

Windows 7

Windows 10

OSX

IOS

Android

Linux


What TCP congestion controls are in use in your cloud, end user
products, and testing regimes?

Reno

Cubic

CDG

Vegas

BBRv1

BBRv2

Other


Is ATF (airtime fairness) in your latest 802.11ac and ax router AP products?


Do you support the Linux APIs for fq_codel for wifi? Presently we only
observe Intel, QCA, and Mediatek support in the Linux mainline. If
not, what are your equivalents for queue management between fast (high
mcs rates) and slow (low mcs rate) stations? Does apple have support
for your equivalents?


Are there APIs and cross compilers to add these features available to
developers? How may they gain access?


[1] https://bufferbloat-and-beyond.net/



-- 
Latest Podcast:
https://www.linkedin.com/feed/update/urn:li:activity:6791014284936785920/

Dave Täht CTO, TekLibre, LLC

             reply	other threads:[~2021-05-18 13:48 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-05-18 13:48 Dave Taht [this message]
2021-05-18 15:16 ` Erik Auerswald

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='CAA93jw4Oi2s=yq89aeFJ+PpzAcZYZ_ivj5pdaqqTCvSMJN3x1Q@mail.gmail.com' \
    --to=dave.taht@gmail.com \
    --cc=bloat@lists.bufferbloat.net \
    /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