Re: [bmwg] [Technical Errata Reported] RFC8239 (6763)

[Leonard Yu <[e-mail]>]

Hi Lucien,

Sorry, I accidentally deleted the entire conversation.

I am still confused about the text. Please let me explain my confusion wrt 3) Measure maximum port pair buffer sizes.:

>From “Ingress port 1 sending line rate to egress port 2”, I know that port 1 and port 2 are a pair.
>From “ingress port 3 sending line rate to egress port 4”, I know that port 3 and port 4 are a pair.
(are my understanding above correct?) At least they are the same as your drawing below.

[Diagram  Description automatically generated]

Then, the text says ” Ingress port N-1 and port N will oversubscribe, at 1% of line
         rate, egress port 2 and port 3, respectively.”. Here, this is where my confusion is. In your drawing, the oversubscribed ports are port 1 and port 2, because they are a pair. But the text says, port 2 and port 3 to be oversubscribed. Port 2 and port 3 are from two different pairs, correct?

They cannot be both correct. Either the text, or the equation is correct.

Best,
Leonard

From: Lucien <[e-mail]>
Date: Wednesday, 1 December 2021 at 22.52
To: Leonard Yu <[e-mail]>
Cc: Jacob Rapp <[e-mail]>, MORTON JR., AL <[e-mail]>, RFC Errata System <[e-mail]>, Sarah B <[e-mail]>, Warren Kumari <[e-mail]>, [e-mail] <[e-mail]>, [e-mail] <[e-mail]>
Subject: Re: [Technical Errata Reported] RFC8239 (6763)
Hi Leonard,

This method depicts iteration to test a port pair. The example in the text uses for example port pair 2 and 3. I did a drawing showing you a different port-pair starting point (1 and 2) instead of (2 and 3). What is important is to know what port-pair is selected and follow the methodology for that test as described in the text.

I suggest engaging in an email conversation if there is further confusion. I am happy to help explain any misunderstandings.

Thank you,

Lucien



On Wed, Dec 1, 2021 at 05:54 Lucien <[e-mail]<mailto:[e-mail]>> wrote:
Iteration 1 and 2 are correct in the original text as well.


On Wed, Dec 1, 2021 at 00:37 Leonard Yu <[e-mail]<mailto:[e-mail]>> wrote:
I was assuming the texts describing Iteration 1 & 2 were incorrect. But from the drawing below, I can see it is the text describing iteration 1 and 2 are incorrect.

I will make a new errara report for to correct the texts for iteration 1 &2
[Text, letter    Description automatically generated]

From: Lucien <[e-mail]<mailto:[e-mail]>>
Date: Tuesday, 30 November 2021 at 20.56
To: MORTON JR., AL <[e-mail]<mailto:[e-mail]>>
Cc: Jacob Rapp <[e-mail]<mailto:[e-mail]>>, RFC Errata System <[e-mail]<mailto:[e-mail]>>, Sarah B <[e-mail]<mailto:[e-mail]>>, Warren Kumari <[e-mail]<mailto:[e-mail]>>, [e-mail]<mailto:[e-mail]> <[e-mail]<mailto:[e-mail]>>, Leonard Yu <[e-mail]<mailto:[e-mail]>>, [e-mail]<mailto:[e-mail]> <[e-mail]<mailto:[e-mail]>>
Subject: Re: [Technical Errata Reported] RFC8239 (6763)
Here is a picture illustrating it, sorry my ascii skills for drawing are not there yet.

[cid:image003.jpg@01D7ED06.33801E20]


On Tue, Nov 30, 2021 at 11:55 MORTON JR., AL <[e-mail]<mailto:[e-mail]>> wrote:
Thanks for your quick reply, Lucien!
Al


From: Lucien <[e-mail]<mailto:[e-mail]>>
Sent: Tuesday, November 30, 2021 2:29 PM
To: RFC Errata System <[e-mail]<mailto:[e-mail]>>
Cc: Jacob Rapp <[e-mail]<mailto:[e-mail]>>; Warren Kumari <[e-mail]<mailto:[e-mail]>>; [e-mail]<mailto:[e-mail]>; MORTON JR., AL <[e-mail]<mailto:[e-mail]>>; Sarah B <[e-mail]<mailto:[e-mail]>>; [e-mail]<mailto:[e-mail]>; [e-mail]<mailto:[e-mail]>
Subject: Re: [Technical Errata Reported] RFC8239 (6763)

Hello,

This errata is incorrect. The original text is correct.

Lucien

On Tue, Nov 30, 2021 at 6:47 AM RFC Errata System <[e-mail]<mailto:[e-mail]>> wrote:
The following errata report has been submitted for RFC8239,
"Data Center Benchmarking Methodology".

--------------------------------------
You may review the report below and at:
https://www.rfc-editor.org/errata/eid6763<https://urldefense.com/v3/__https:/www.rfc-editor.org/errata/eid6763__;!!BhdT!wScd1KMc51KUWqw0VGpG0Koz6W11ctc3VWC0cf5ClMcedcAYPseHP0RwShhp$>

--------------------------------------
Type: Technical
Reported by: Leonard Yu <[e-mail]<mailto:[e-mail]>>

Section: 3.2

Original Text
-------------
Last iteration: Ingress port 1 sending line rate to egress
port 2, ingress port 3 sending line rate to egress port 4, etc.
Ingress port N-1 and port N will oversubscribe, at 1% of line
rate, egress port N-3 and port N-2, respectively. Measure the
buffer size value by multiplying the number of extra frames
sent by the frame size for each egress port.

Corrected Text
--------------
Last iteration: Ingress port 1 sending line rate to egress
port 2, ingress port 3 sending line rate to egress port 4, etc.
Ingress port N-1 and port N will oversubscribe, at 1% of line
rate, egress port N-4 and port N-3, respectively. Measure the
buffer size value by multiplying the number of extra frames
sent by the frame size for each egress port.

Notes
-----
If
#1, 1->2, 3->4, 5->6, 7->8, ... and N-1->2, N->3
#2, 1->2, 3->4, 5->6, 7->8, ... and N-1->4, N->5

Then
#last, 1->2, 3->4, 5->6, 7->8, ... and N-1->N-4, N->N-3

Otherwise, the general equation won't satisfy #1 and #2

Instructions:
-------------
This erratum is currently posted as "Reported". If necessary, please
use "Reply All" to discuss whether it should be verified or
rejected. When a decision is reached, the verifying party
can log in to change the status and edit the report, if necessary.

--------------------------------------
RFC8239 (draft-ietf-bmwg-dcbench-methodology-18)
--------------------------------------
Title               : Data Center Benchmarking Methodology
Publication Date    : August 2017
Author(s)           : L. Avramov, J. Rapp
Category            : INFORMATIONAL
Source              : Benchmarking Methodology
Area                : Operations and Management
Stream              : IETF
Verifying Party     : IESG