[Codel] Comparing Linux qdiscs in lab conditions (paper)

[Codel] Comparing Linux qdiscs in lab conditions (paper)

[Toke Høiland-Jørgensen]

[-- Attachment #1: Type: text/plain, Size: 774 bytes --]

Hello

I've recently written and defended this university project report on
bufferbloat as part of my masters programme:

http://rudar.ruc.dk/handle/1800/9322 (or
http://akira.ruc.dk/~tohojo/bufferbloat/bufferbloat-final.pdf if the
first link doesn't work).

The paper compares four different Linux qdiscs in a simple lab setup,
and demonstrates bufferbloat, and the mitigation provided by fq_codel,
quite well. The first couple of sections on how TCP and the Linux
network stack work are probably known material to most people here, so
feel free to skip to the experimental setup and results in sections 5
and 6.

I hope you will find this interesting; feel free to comment and/or ask
questions. :)

-Toke

-- 
Toke Høiland-Jørgensen
toke@toke.dk

[-- Attachment #2: Type: application/pgp-signature, Size: 489 bytes --]

Re: [Codel] Comparing Linux qdiscs in lab conditions (paper)

[Stephen Hemminger]

[-- Attachment #1: Type: text/plain, Size: 1120 bytes --]

On Fri, 18 Jan 2013 00:50:22 +0100
Toke Høiland-Jørgensen <toke@toke.dk> wrote:

> Hello
> 
> I've recently written and defended this university project report on
> bufferbloat as part of my masters programme:
> 
> http://rudar.ruc.dk/handle/1800/9322 (or
> http://akira.ruc.dk/~tohojo/bufferbloat/bufferbloat-final.pdf if the
> first link doesn't work).
> 
> The paper compares four different Linux qdiscs in a simple lab setup,
> and demonstrates bufferbloat, and the mitigation provided by fq_codel,
> quite well. The first couple of sections on how TCP and the Linux
> network stack work are probably known material to most people here, so
> feel free to skip to the experimental setup and results in sections 5
> and 6.
> 
> I hope you will find this interesting; feel free to comment and/or ask
> questions. :)
> 
> -Toke
> 

I am doing a demo at LCA in two weeks. Your graphs are the clearest examples
of different qdisc, and mesh nicely with my fluid demos. If it is okay,
could you send me the originals for the Figure 5 graphs, I will use them with attribution
during the summary.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [Codel] Comparing Linux qdiscs in lab conditions (paper)

[Toke Høiland-Jørgensen]

[-- Attachment #1.1: Type: text/plain, Size: 654 bytes --]

Stephen Hemminger <stephen@networkplumber.org> writes:

> I am doing a demo at LCA in two weeks. Your graphs are the clearest
> examples of different qdisc, and mesh nicely with my fluid demos. If
> it is okay, could you send me the originals for the Figure 5 graphs, I
> will use them with attribution during the summary.

Sure, go ahead (attached). :)

They are also available here: http://akira.ruc.dk/~tohojo/bufferbloat/
along with the raw data sets which can be fed into netperf-wrapper
(https://github.com/tohojo/netperf-wrapper) to generate various plots
in various formats.

-Toke

-- 
Toke Høiland-Jørgensen
toke@toke.dk


[-- Attachment #1.2: tcp_bidirectional-codel-totals.pdf --]
[-- Type: application/pdf, Size: 20066 bytes --]

[-- Attachment #1.3: tcp_bidirectional-fq_codel-totals.pdf --]
[-- Type: application/pdf, Size: 19800 bytes --]

[-- Attachment #1.4: tcp_bidirectional-pfifo_fast-totals.pdf --]
[-- Type: application/pdf, Size: 20122 bytes --]

[-- Attachment #1.5: tcp_bidirectional-sfq-totals.pdf --]
[-- Type: application/pdf, Size: 17475 bytes --]

[-- Attachment #1.6: tcp_bidirectional-ping_cdf.pdf --]
[-- Type: application/pdf, Size: 21648 bytes --]

[-- Attachment #2: Type: application/pgp-signature, Size: 489 bytes --]