tsvwg, aqm, tcpm, tcpPrague mailing lists,
A few people have been working away to specify and document all the
aspects of the new Low Latency, Low Loss, Scalable throughput (L4S)
service, which held a successful BoF in Berlin. As the decision was to
try to work across multiple WGs, I thought it would be useful to give ...
... the status collected over all activities and WGs. I've included
stuff that has wider applicability but is also needed for L4S. Click on
the headings to take you to the page where all these links are collected
together.
Source Code <https://riteproject.eu/dctth/#code>
* Dual Queue Coupled AQM
o With Curvy RED for Linux (access available shortly)
o With PI2 for Linux <https://github.com/olgabo/dualpi2> [*UPDATED*]
* Data Centre TCP (DCTCP) for
o Linux (in the mainline kernel <https://www.kernel.org/>)
o FreeBSD patch
<http://lists.freebsd.org/pipermail/freebsd-net/2014-February/037915.html>
o ns2 patch <http://simula.stanford.edu/%7Ealizade/Site/DCTCP.html>.
* Accurate ECN TCP Feedback for Linux
<https://github.com/mirjak/linux-accecn/> [*COMPLETED*, but not
fully tested]
*IETF specs* <https://riteproject.eu/dctth/#ietf-specs>
As the architecture explains, there are three main parts to
standardisation: the identifier (#1), the network algorithms (#2) and
the host algorithms (#3):
* Low Latency, Low Loss, Scalable Throughput (L4S) Internet Service:
Architecture and Problem Statement
<draft-briscoe-tsvwg-l4s-arch
<https://tools.ietf.org/html/draft-briscoe-tsvwg-l4s-arch>> [*NEW*]
1. A proposed new identifier for Low Latency, Low Loss, Scalable
throughput (L4S) packets
<draft-briscoe-tsvwg-ecn-l4s-id
<https://tools.ietf.org/html/draft-briscoe-tsvwg-ecn-l4s-id>>
[*UPDATED to reflect the proposed process*]
and the IETF’s enabling draft to make way for this
<draft-black-tsvwg-ecn-experimentation
<https://tools.ietf.org/html/draft-black-tsvwg-ecn-experimentation-02.txthttps://tools.ietf.org/html/draft-black-tsvwg-ecn-experimentation>>
[*NEW*]
2. Then network operators can deploy a new simple active queue
management algorithm, that complies with the few constraints
specified here:
<draft-briscoe-tsvwg-aqm-dualq-coupled
<https://tools.ietf.org/html/draft-briscoe-tsvwg-aqm-dualq-coupled>>
Example algorithms are given in appendices. [*PI2 pseudocode added
to make 2 example**appendices*]
3. And host developers can deploy new scalable TCP algorithms, e.g.
Data Centre TCP (DCTCP):
<draft-ietf-tcpm-dctcp
<https://tools.ietf.org/html/draft-ietf-tcpm-dctcp>>
Without steps #2 & #3, scalable TCPs are too aggressive to coexist
with classic TCPs like Reno and Cubic, so DCTCP was initially
confined to controlled networks like Data Centres, where all classic
traffic sources could be upgraded (or isolated). To be safe over the
public Internet, scalable TCP algorithms will need to comply with
the list of requirements agreed at an informally convened meeting in
Prague that have become know as the TCP Prague requirements
<https://mailarchive.ietf.org/arch/msg/tcpprague/mwWncQg3egPd15FItYWiEvRDrvA>:
* One of these requirements is accurate ECN feedback, which the IETF
has recognised will need an update to the TCP protocol, and it has
stated the requirements a solution should comply with:
<RFC7560 <http://tools.ietf.org/html/rfc7560>>
A candidate protocol to satisfy these requirements has been
developed and adopted by the IETF:
<draft-ietf-tcpm-accurate-ecn
<https://tools.ietf.org/html/draft-ietf-tcpm-accurate-ecn>> [*UPDATED*]
* The following drafts address other specific TCP Prague requirements:
Adding ECN to TCP control packets:
<draft-bagnulo-tcpm-generalized-ecn
<https://tools.ietf.org/html/draft-bagnulo-tcpm-generalized-ecn>>
[*UPDATED*]
Recommendations for increasing TCP performance in low RTT networks:
<draft-bagnulo-tcpm-tcp-low-rtt
<https://tools.ietf.org/html/draft-bagnulo-tcpm-tcp-low-rtt>>
Papers <https://riteproject.eu/dctth/#papers>
* Article in the IETF Journal describing the Demo in Bits-N-Bites at
the IETF in Prague, July 2015.
“Ultra-Low Delay for All
<http://www.internetsociety.org/publications/ietf-journal-november-2015/ultra-low-delay-for-all>”
IETF Journal, Nov 2015.
* Description of an**ambitious demo of five low latency apps on one
broadband line, 4 of which were also high throughput:
“Ultra-Low Delay for All: Live Experience, Live Analysis
<https://riteproject.files.wordpress.com/2015/10/uld4all-demo_mmsys.pdf>“,
Proc. ACM Multimedia Systems; Demo Session (May 2016).
* “/*PI^2 : A Linearized AQM for both Classic and Scalable TCP*/
<https://riteproject.files.wordpress.com/2015/10/pi2_conext.pdf>,”
Proc. ACM CONEXT 2016 (To appear Dec 2016). [*NEW*]
*
Finally, a pathetic apology is in order: We prepared a [*NEW*] paper
for a top conference giving the full L4S story, with a massive set
of testbed-based evaluations, with a wide range of link rates, RTTs,
mixtures of numbers of flows in each queue, etc. etc. But we
screwed up, and it got rejected without even being read - we
exceeded the page limit :(
We don't want to publish it on the Web until we've tried again. But
if anyone wants a private copy, pls just ask.
Otherwise, the following older paper is still available, but not as
comprehensive or well-honed:
Pre-print of paper explaining why scalable TCP algorithms solve the
problem (in plain English and maths), how the Dual Queue Coupled AQM
algorithm works, and recording the results of extensive testbed
experiments.
“`/*Data Centre to the Home’: Ultra-Low Latency for All*/
<http://www.bobbriscoe.net/projects/latency/dctth_preprint.pdf>” (2015)
Thank you to everyone involved so far.
Cheers
Bob
--
________________________________________________________________
Bob Briscoe http://bobbriscoe.net/
_______________________________________________
aqm mailing list
aqm@ietf.org
https://www.ietf.org/mailman/listinfo/aqm