Re: [tsvwg] start of WGLC on L4S drafts

[Bob Briscoe <[e-mail]>]

Ermin,

Thanks for the comments on the drafts during WGLC. See [BB] inline for 
responses. Where there is no response, I just took the comment as read.

On 16/08/2021 08:59, Ermin Sakic wrote:
> Hi Wesley, all,
>
> L4S proposals are an important step towards enabling low-latency transmissions for remote control, AR/VR, and online gaming applications.
> I support progressing the drafts to RFC status.
>
> I have attached a few comments below after re-reading the latest versions of the three drafts.
> My apologies if some of the comments came up in discussion earlier.
>
> Best wishes,
> Ermin Sakic, NVIDIA
>
> - draft-ietf-tsvwg-l4s-arch-10:
>          - Figure 1: Is there a specific reason why the “mark/drop” action is not denoted for L4S queue compared to “mark”-only? Section 5.1 of draft-ietf-tsvwg-ecn-l4s-id-19 states that “…L4S marking treatment MUST begin applying drop to L4S traffic until the overload episode has subsided…”.

[BB] On balance I think it's best for a high level figure like this to 
show only the behaviour expected under normal circumstances. This is why 
it doesn't show drop as well, because Not-ECT traffic is not classified 
into the L queue.

But, you're right that it's important to emphasize at every opportunity 
that the L queue has to apply drop if overloaded. So I've added the 
following to the text describing the figure:
     "If either queue becomes persistently overloaded, ECN marking is 
disabled, as recommended in Section 7 of [RFC3168] and Section 4.2.1 of 
[RFC7567]. Then both queues introduce the same level of drop (not shown 
in the figure)."


>          - Section 8.2: “…if it is leads…” -> “…if it leads…”
> - draft-ietf-tsvwg-ecn-l4s-id-19:
>          - Section 1 Introduction: “…responses to them result..” -> “…responses to markings result…”

[BB] Yes, 'them' wasn't clear, but this highlights another problem - 
it's not clear what the "control signals" are. So rather than use a 
different word ("markings"), how about referring back to 'these 
signals', and adding 'ECN' to the first occurrence.
     "...the much more frequent ECN control signals, and the finer 
responses to these signals, result in..."

>          - Section 1.1: “The explanation is summarised without the maths in the L4S architecture document” -> I suggest referencing to the specific section number of draft-ietf-tsvwg-l4s-arch here.
>          - Section 7.1: “TCP. QUIC…” -> “TCP, QUIC…”
>          - Appendix B: “tranport-layer” -> “transport-layer”
> - draft-ietf-tsvwg-aqm-dualq-coupled-16:
>          - Figure 1: As mentioned above, I’d expect to see an optional drop as the outcome of forwarding to the L traffic queue in addition to the “ECN marking” action.

Again, I'd rather not complicate the "opening lesson of the tutorial" 
with exceptions that come later. But it would be good to highlight that 
exceptions will be coming later. So instead of adding to the diagram, 
I've added to the first para of the section, which already warns of one 
case where the diagram might not be authoritative:

"Also, the schematic only illustrates operation under normally expected 
circumstances; behaviour under overload or with operator-specific 
classifiers is deferred to Section 2.5.1.1"

These edits will appear alongside those for the other recent comments, 
before the draft deadline on Monday.

And I've ack'd your contribution in all three drafts. Thank you again.


Bob

>
> -----Original Message-----
> From: tsvwg <[e-mail]> On Behalf Of Wesley Eddy
> Sent: Thursday, 29 July 2021 18:18
> To: [e-mail]
> Subject: [tsvwg] start of WGLC on L4S drafts
>
> External email: Use caution opening links or attachments
>
>
> This message is starting a combined working group last call on 3 of the L4S drafts:
>
> - Architecture: https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdatatracker.ietf.org%2Fdoc%2Fdraft-ietf-tsvwg-l4s-arch%2F&amp;data=04%7C01%7Cesakic%40nvidia.com%7C911e34a8568a4c43d67c08d952ac82f5%7C43083d15727340c1b7db39efd9ccc17a%7C0%7C0%7C637631724620979753%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&amp;sdata=a7V1lmfJi%2Fbs2H2mZIr3QhveYsPj3QzPguBFDWsy%2BC0%3D&amp;reserved=0
>
> - DualQ:
> https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdatatracker.ietf.org%2Fdoc%2Fdraft-ietf-tsvwg-aqm-dualq-coupled%2F&amp;data=04%7C01%7Cesakic%40nvidia.com%7C911e34a8568a4c43d67c08d952ac82f5%7C43083d15727340c1b7db39efd9ccc17a%7C0%7C0%7C637631724620979753%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&amp;sdata=PwO0BFTNKv2c4ZbGoQZEvxfLg7NIG9mw8oWtNqdwsjs%3D&amp;reserved=0
>
> - ECN ID: https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdatatracker.ietf.org%2Fdoc%2Fdraft-ietf-tsvwg-ecn-l4s-id%2F&amp;data=04%7C01%7Cesakic%40nvidia.com%7C911e34a8568a4c43d67c08d952ac82f5%7C43083d15727340c1b7db39efd9ccc17a%7C0%7C0%7C637631724620979753%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&amp;sdata=QpBxzuleCeXpxzoBR4oEJWANLk7v1Vvo%2Ft4H8d3kZz0%3D&amp;reserved=0
>
> The WGLC will last through 4 weeks from today, and then we'll see what to do next.  Please submit any comments you have on these to the TSVWG list in that timeframe.
>
> The chairs are considering a possible virtual interim following the close in order to work through feedback received.
>
> The work on the L4S operational guidance draft is continuing in parallel, but that draft is not being last called yet.
>
>
> -----------------------------------------------------------------------------------
> NVIDIA GmbH
> Wuerselen
> Amtsgericht Aachen
> HRB 8361
> Managing Director: Karen Theresa Burns
>
> -----------------------------------------------------------------------------------
> This email message is for the sole use of the intended recipient(s) and may contain
> confidential information.  Any unauthorized review, use, disclosure or distribution
> is prohibited.  If you are not the intended recipient, please contact the sender by
> reply email and destroy all copies of the original message.
> -----------------------------------------------------------------------------------

-- 
________________________________________________________________
Bob Briscoe                               http://bobbriscoe.net/