Re: [tcpm] [EXTERNAL] Re: WGLC for draft-ietf-tcpm-hystartplusplus-04

Hi Huang,

Thanks for updating draft.
I found some parts has been updated (thank you!)
But, as far as I read, I think the following part I mentioned has not been
changed, I'm curious if there're some reasons for it.

Section 4.2: "For CSS rounds where N_RTT_SAMPLE RTT samples have been
obtained,"

Also, from my point of view, Section 5 is still not very clear what was
compared with hystart++.
I am thinking that it might be good to clarify this point if there's no
harm for it.

Thanks,
--
Yoshi

On Mon, May 16, 2022 at 2:29 PM Yi Huang <huanyi@microsoft.com> wrote:

> Hi Yoshifumi,
>
> Thanks for the review!
>
>
>    1. We will update the text as you suggested.
>    2. We were comparing Hystart++ with conventional slow-start. We have
>    never implemented the original version of Hystart and I agree the data we
>    have right now is a bit weak. We will consider adding some lab results to
>    show that Hystart++ with CSS works better than Hystart++ without CSS in
>    networks with latency jitters.
>
> Yi
>
> ------------------------------
> *From:* tcpm <tcpm-bounces@ietf.org> on behalf of Yoshifumi Nishida <
> nsd.ietf@gmail.com>
> *Sent:* Sunday, May 1, 2022 11:05 PM
> *To:* tcpm IETF list <tcpm@ietf.org>
> *Subject:* [EXTERNAL] Re: [tcpm] WGLC for
> draft-ietf-tcpm-hystartplusplus-04
>
> Hello,
> As an individual participant,  I think the draft is mostly ready, but I
> have the following comments.
>
> 1: Section 4.2:
>     according to the pseudo code, rttSampleCount can exceed N_RTT_SAMPLE.
>     So, I personally prefer to update the following sentences to mention
> this point more explicitly like:
>    "For rounds where N_RTT_SAMPLE RTT samples have been obtained and
> currentRoundMinRTT and lastRoundMinRTT are valid"
>           ->
>    "For rounds where at least N_RTT_SAMPLE RTT samples have been
> obtained..."
>
>    "For CSS rounds where N_RTT_SAMPLE RTT samples have been obtained.."
>           ->
>    "For CSS rounds where at least N_RTT_SAMPLE RTT samples have been
> obtained.."
>
> 2: Section 5:
>     "Hystart++ reduces byte retransmitted by 50% and retransmission
> timeouts by 36%."
>     "In an A/B test for Hystart++ draft.."
>
>      It's not very clear what was compared with hystart++ here. Is it
> conventional slow-start or hystart?
>      If it's hystart, it seems to me that reducing retransmission and
> retransmission timeout is a bit counter-intuitive as hystart++ can gain
> higher throughput than hystart in my understanding.
>      If it's conventional slow-start, the presentation of the results is a
> bit weak as the motivation of hystart++ is to improve hystart.
>
> Thanks,
> --
> Yoshi
>
>
> On Wed, Apr 13, 2022 at 1:19 PM Michael Tuexen <tuexen@fh-muenster.de>
> wrote:
>
> Dear all,
>
> this e-mail starts the working group last call for
> draft-ietf-tcpm-hystartplusplus-04.
>
> The WGLC runs until Friday, May 6th 2022.
>
> Please send any comments, including indications to support this document,
> to the TCMP mailing list by then.
>
> The ID is available at
> https://www.ietf.org/archive/id/draft-ietf-tcpm-hystartplusplus-04.html
> <https://nam06.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.ietf.org%2Farchive%2Fid%2Fdraft-ietf-tcpm-hystartplusplus-04.html&data=05%7C01%7Chuanyi%40microsoft.com%7C15a4c3c1fb334b3fcfe308da2c01cb93%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C637870683603858503%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C2000%7C%7C%7C&sdata=rJ%2B0mN6IulVDmdcQbgZ6hov3hWI7bRwhZhOj6CrrHFA%3D&reserved=0>
>
> Best regards
> Michael
>
>
> _______________________________________________
> tcpm mailing list
> tcpm@ietf.org
> https://www.ietf.org/mailman/listinfo/tcpm
> <https://nam06.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.ietf.org%2Fmailman%2Flistinfo%2Ftcpm&data=05%7C01%7Chuanyi%40microsoft.com%7C15a4c3c1fb334b3fcfe308da2c01cb93%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C637870683603858503%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C2000%7C%7C%7C&sdata=C1VbtrGBlPIx5qEnry%2F4tO6KSsccfU8KTW7AwQazEcY%3D&reserved=0>
>
>