[tsvwg] New SCE-related I-Ds

[Jonathan Morton <[e-mail]>]

We've just submitted three new Internet Drafts related to our ongoing work on SCE.  Rolling these up into a single notification…

> Name:		draft-morton-tsvwg-sce
> Revision:	00
> Title:		The Some Congestion Experienced ECN Codepoint
> Document date:	2019-07-03
> Group:		Individual Submission
> Pages:		9
> URL:            https://www.ietf.org/internet-drafts/draft-morton-tsvwg-sce-00.txt
> Status:         https://datatracker.ietf.org/doc/draft-morton-tsvwg-sce/
> Htmlized:       https://tools.ietf.org/html/draft-morton-tsvwg-sce-00
> Htmlized:       https://datatracker.ietf.org/doc/html/draft-morton-tsvwg-sce
> 
> 
> Abstract:
>   This memo reclassifies ECT(1) to be an early notification of
>   congestion on ECT(0) marked packets, which can be used by AQM
>   algorithms and transports as an earlier signal of congestion than CE.
>   It is a simple, transparent, and backward compatible upgrade to
>   existing IETF-approved AQMs, RFC3168, and nearly all congestion
>   control algorithms.

Although this appears to be a -00, that's because it was submitted under a new filename to better match IETF conventions.  It is a major revision of draft-morton-taht-tsvwg-sce-00, addressing in particular the interaction with single-queue AQMs supporting SCE, and also bringing the text into line with the current state of our reference implementation.

> Name:		draft-morton-tsvwg-lightweight-fair-queueing
> Revision:	00
> Title:		Lightweight Fair Queueing
> Document date:	2019-07-03
> Group:		Individual Submission
> Pages:		10
> URL:            https://www.ietf.org/internet-drafts/draft-morton-tsvwg-lightweight-fair-queueing-00.txt
> Status:         https://datatracker.ietf.org/doc/draft-morton-tsvwg-lightweight-fair-queueing/
> Htmlized:       https://tools.ietf.org/html/draft-morton-tsvwg-lightweight-fair-queueing-00
> Htmlized:       https://datatracker.ietf.org/doc/html/draft-morton-tsvwg-lightweight-fair-queueing
> 
> 
> Abstract:
>   This note presents Lightweight Fair Queueing (LFQ), a fair queueing
>   algorithm with a small code footprint, low memory requirements, no
>   multiply operations, only two physical queues, and only one set of
>   AQM state.  LFQ provides throughput fairness, sparse flow
>   prioritization and ordering guarantees, making it suitable for a
>   mixture of traffic flow types.

Although we found that SCE can be made to work acceptably well with single-queue AQMs, we felt that a middle ground between single-queue AQM and full FQ was relevant to high-capacity networks, in which a dual-FIFO queue discipline seems to be the maximum acceptable.  We therefore developed LFQ to approach the favourable flow-isolation performance of DRR++ with considerably reduced implementation complexity.  In turn, this allows transports employing high-fidelity ECN methods to coexist with conventional transports on the same link, without requiring an explicit identifier, nor heuristic fallbacks at the endpoints.

Presently we have a working offline simulation of LFQ, and we hope to have a practical implementation for real-network experiments by Montreal.  Ideas for efficiently implementing multi-level fairness are also being explored, and may appear in a later version of this draft.

> Name:		draft-heist-tsvwg-sce-one-and-two-flow-tests
> Revision:	00
> Title:		Some Congestion Experienced One and Two-Flow Tests
> Document date:	2019-07-03
> Group:		Individual Submission
> Pages:		66
> URL:            https://www.ietf.org/internet-drafts/draft-heist-tsvwg-sce-one-and-two-flow-tests-00.txt
> Status:         https://datatracker.ietf.org/doc/draft-heist-tsvwg-sce-one-and-two-flow-tests/
> Htmlized:       https://tools.ietf.org/html/draft-heist-tsvwg-sce-one-and-two-flow-tests-00
> Htmlized:       https://datatracker.ietf.org/doc/html/draft-heist-tsvwg-sce-one-and-two-flow-tests
> 
> 
> Abstract:
>   This note presents one and two-flow test results for the SCE (Some
>   Congestion Experienced) reference implementation.  These tests are
>   not intended to be a comprehensive real-world evaluation of SCE, but
>   an illustration of SCE's influence on basic TCP metrics in a
>   controlled environment.

To demonstrate our progress, we are also publishing initial test results from our reference implementation.

Constructive commentary is welcomed.

 - Jonathan Morton