Re: [Rdma-cc-interest] IETF-106 side meeting actionable requests

"Roni Even (A)" <roni.even@huawei.com> Tue, 19 November 2019 08:33 UTC

From: "Roni Even (A)" <roni.even@huawei.com>
To: Paul Congdon <paul.congdon@tallac.com>, "rdma-cc-interest@ietf.org" <rdma-cc-interest@ietf.org>
Thread-Topic: [Rdma-cc-interest] IETF-106 side meeting actionable requests
Thread-Index: AQHVnrEz8H7qNUBqnU6ejEkpGb/v6KeSKgZw
Date: Tue, 19 Nov 2019 08:32:52 +0000
Message-ID: <6E58094ECC8D8344914996DAD28F1CCD23DC67EC@dggemm526-mbx.china.huawei.com>
References: <CAAMqZPtQd4Aq_UA0FGkL64PRdmLNJnQVHaKPAVgtuO++1ZVpeQ@mail.gmail.com>
In-Reply-To: <CAAMqZPtQd4Aq_UA0FGkL64PRdmLNJnQVHaKPAVgtuO++1ZVpeQ@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
Content-Type: multipart/alternative; boundary="_000_6E58094ECC8D8344914996DAD28F1CCD23DC67ECdggemm526mbxchi_"
MIME-Version: 1.0
Archived-At: <https://mailarchive.ietf.org/arch/msg/rdma-cc-interest/_ZaWyeL0g0mO6xa4L85_z15QI-8>
Subject: Re: [Rdma-cc-interest] IETF-106 side meeting actionable requests
Precedence: list

Hi Paul,
One more action item for fast congestion.
List the parameters that were used for the test for example threshold time for sending control message.
Roni

From: Rdma-cc-interest [mailto:rdma-cc-interest-bounces@ietf.org] On Behalf Of Paul Congdon
Sent: Tuesday, November 19, 2019 10:13 AM
To: rdma-cc-interest@ietf.org
Subject: [Rdma-cc-interest] IETF-106 side meeting actionable requests

Thank you for attending the side meeting at IETF-106.  I'll be sending the complete set of minutes out in the future, for now I wanted to confirm the set of actionable requests I heard during the draft reviews.  Please let me know if I missed anything.

Summary of actionable requests from the side meeting:  NOTE: these are not detailed minutes.

1.       Fast Congestion management for Data Centers – Roni Even – 20 mins

•         https://tools.ietf.org/html/draft-even-iccrg-dc-fast-congestion-00
Actions:

a.       Perform tests with more complex topology – not just a single switch

b.       Distinguish the solution clearly from ICMP Source Quench and understand the previous issues with that approach

c.       Make it clear how the proposed solution works in an overlay/tunneled environment

2.       An Open Congestion Control Architecture for high performance fabrics - Yan Zhuang – 15 mins

•         https://datatracker.ietf.org/doc/draft-zhuang-tsvwg-open-cc-architecture/
Actions:

a.       Please provide the details of the API

3.       Artificial Intelligence (AI) based ECN adaptive reconfiguration for datacenter networks - Yan Zhuang – 15 mins

•         https://datatracker.ietf.org/doc/draft-zhuang-tsvwg-ai-ecn-for-dcn
Actions:

a.       The analysis should be re-run using a modern AQM – not RED

b.       Provide details about the model’s definition of ‘normal’ behavior

4.       The impact of mixing TCP and RoCEv2 – Yolanda Yu - 10 mins
Actions:

a.       Provide excruciating detail about the configuration of the test environment

[Rdma-cc-interest] IETF-106 side meeting actionab… Paul Congdon
Re: [Rdma-cc-interest] IETF-106 side meeting acti… Roni Even (A)