Return-Path: X-Original-To: idr@ietfa.amsl.com Delivered-To: idr@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 93815C1D4A70; Thu, 4 Jul 2024 20:00:29 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -6.905 X-Spam-Level: X-Spam-Status: No, score=-6.905 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_ZEN_BLOCKED_OPENDNS=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, T_SCC_BODY_TEXT_LINE=-0.01, URIBL_BLOCKED=0.001, URIBL_DBL_BLOCKED_OPENDNS=0.001, URIBL_ZEN_BLOCKED_OPENDNS=0.001] autolearn=unavailable autolearn_force=no Received: from mail.ietf.org ([50.223.129.194]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id GiDoAJnGYcU0; Thu, 4 Jul 2024 20:00:25 -0700 (PDT) Received: from mail-m49198.qiye.163.com (mail-m49198.qiye.163.com [45.254.49.198]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id A4C15C1CAF4A; Thu, 4 Jul 2024 20:00:23 -0700 (PDT) Received: from LAPTOP09T7970K (unknown [219.142.69.75]) by smtp.qiye.163.com (Hmail) with ESMTPA id DD12E7E01C5; Fri, 5 Jul 2024 11:00:14 +0800 (CST) From: "Aijun Wang" To: =?utf-8?B?J+mcjeaci+mjnic=?= , , References: In-Reply-To: Date: Fri, 5 Jul 2024 11:00:14 +0800 Message-ID: <000501dace87$75a315e0$60e941a0$@tsinghua.org.cn> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0006_01DACECA.83C76750" X-Mailer: Microsoft Outlook 16.0 Thread-Index: AQIR1ITPc6GYbMskLLmLageatLZx17F5XtqQ Content-Language: zh-cn X-HM-Spam-Status: e1kfGhgUHx5ZQUpXWQgPGg8OCBgUHx5ZQUlOS1dZFg8aDwILHllBWSg2Ly tZV1koWUFKTEtLSjdXWS1ZQUlXWQ8JGhUIEh9ZQVlCGkweVktDHkofSE8fQxkaGFYeHw5VEwETFh oSFyQUDg9ZV1kYEgtZQVlJSkJVSk9JVU1CVUxOWVdZFhoPEhUdFFlBWU9LSFVKS0lPT09IVUpLS1 VKQktLWQY+ X-HM-Tid: 0a9080d5d22903a2kunmdd12e7e01c5 X-HM-MType: 10 X-HM-Sender-Digest: e1kMHhlZQR0aFwgeV1kSHx4VD1lBWUc6NjI6Kjo5LjNDTlZKDhNNFiI4 Dw0KFD9VSlVKTElLSk9DT0pOTExOVTMWGhIXVQwaFRwaEhEOFTsPCBIVHBMOGlUUCRxVGBVFWVdZ EgtZQVlJSkJVSk9JVU1CVUxOWVdZCAFZQUlISEtDNwY+ Message-ID-Hash: LO4FMF4KR3KEL55FBAO7X7YEOJXJMX2W X-Message-ID-Hash: LO4FMF4KR3KEL55FBAO7X7YEOJXJMX2W X-MailFrom: wangaijun@tsinghua.org.cn X-Mailman-Rule-Misses: dmarc-mitigation; no-senders; approved; emergency; loop; banned-address; member-moderation; header-match-idr.ietf.org-0; nonmember-moderation; administrivia; implicit-dest; max-recipients; max-size; news-moderation; no-subject; digests; suspicious-header X-Mailman-Version: 3.3.9rc4 Precedence: list Subject: =?utf-8?q?=5BIdr=5D_=E7=AD=94=E5=A4=8D=3A_=5BRTGWG_=26_IDR=5D_Discussing_the?= =?utf-8?q?_Open_Scheduled_fabric_of_AI_Network?= List-Id: Inter-Domain Routing Archived-At: List-Archive: List-Help: List-Owner: List-Post: List-Subscribe: List-Unsubscribe: This is a multipart message in MIME format. ------=_NextPart_000_0006_01DACECA.83C76750 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Hi, Pengfei: =20 After reading your proposals, I have the following questions: =20 1. In your osf-framework, how you assure the sorting, fragmentation = or aggregation of packet of one flow at the Ingress/Egress device does = not lead to forwarding latency(also large required memory storage on = these devices)? 2. For your congestion control proposal(credit-based flow control), = what=E2=80=99s the difference from the traditional RSVP technology? 3. For your rapid link failure switchover in OSF, the main point is = that the involvement of OSF-MS component that act as the central control = point. Is it right? =20 Where is the OSF-MS component in your architecture that described in = section = https://datatracker.ietf.org/doc/html/draft-hcl-rtgwg-osf-framework-00#se= ction-2.2 =20 4. And, in = https://datatracker.ietf.org/doc/html/draft-hcl-rtgwg-osf-framework-00#se= ction-3, the document states that OSF-MS components can also operate in = a distributed process mode, or a hybrid mode,=20 how do you achieve the rapid link failure switchover as AI applications = required in these two modes? =20 Best Regards =20 Aijun Wang China Telecom =20 =E5=8F=91=E4=BB=B6=E4=BA=BA: forwardingalgorithm@ietf.org = [mailto:forwardingalgorithm@ietf.org] =E4=BB=A3=E8=A1=A8 = =E9=9C=8D=E6=9C=8B=E9=A3=9E =E5=8F=91=E9=80=81=E6=97=B6=E9=97=B4: 2024=E5=B9=B46=E6=9C=8830=E6=97=A5 = 12:57 =E6=94=B6=E4=BB=B6=E4=BA=BA: idr@ietf.org; rtgwg@ietf.org =E4=B8=BB=E9=A2=98: [Idr] [RTGWG & IDR] Discussing the Open Scheduled = fabric of AI Network =20 Hi Wg, =20 Link: = = https://datatracker.ietf.org/doc/html/draft-hcl-rtgwg-ai-network-problem-= 00 = = https://datatracker.ietf.org/doc/html/draft-hcl-rtgwg-osf-framework-00 = = https://datatracker.ietf.org/doc/html/draft-hcl-idr-extend-tunnel-egress-= point-00 =20 These documents describe the main disturbances caused by AI applications = to the network and propose solutions based on open schedulable network = architecture developed to address these issues. It primarily focuses on the following aspects: =20 1. Open Scheduled Fabric An open network architecture that resolves hash issues through dynamic = scheduling based on network measurements. =20 2. Bgp Extension for Tunnel Egress Point A method for implementing the OSF network architecture involves using = BGP uniformly in the control plane, extending the data plane as a new = tunnel encapsulation, and providing the necessary control information = transmission for the data plane. =20 3. Extending the BGP protocol to carry Tunnel Egress Point information = within the community attribute. =20 We welcome any comments and discussions. =20 Thank you! =20 Thanks, Pengfei ------=_NextPart_000_0006_01DACECA.83C76750 Content-Type: text/html; charset="utf-8" Content-Transfer-Encoding: quoted-printable

H= i, Pengfei:

<= o:p> 

A= fter reading your proposals, I have the following = questions:

<= o:p> 

<= span style=3D'mso-list:Ignore'>1.     I= n your osf-framework, how you assure the sorting, fragmentation or = aggregation of packet of one flow at the Ingress/Egress device does not = lead to forwarding latency(also large required memory storage on these = devices)?

<= span style=3D'mso-list:Ignore'>2.     F= or your congestion control proposal(credit-based flow control), = what=E2=80=99s the difference from the traditional RSVP = technology?

3.  F= or your rapid link failure switchover in OSF, the main point is that the = involvement of OSF-MS component that act as the central control point. = Is it right? =C2=A0

W= here is the OSF-MS component in your architecture that described in = section https://datatracker.ietf.org/doc/html/draft-hcl-rtgwg-= osf-framework-00#section-2.2

 

<= span style=3D'mso-list:Ignore'>4.     A= nd, in https://datatracker.ietf.org/doc/html/draft-hcl-rtgwg-os= f-framework-00#section-3, the document states that OSF-MS components = can also operate in a distributed process mode, or a hybrid mode, =

h= ow do you achieve the rapid link failure switchover as AI applications = required in these two modes?

<= o:p> 

B= est Regards

<= o:p> 

A= ijun Wang

C= hina Telecom<= o:p>

<= o:p> 

=E5=8F=91=E4=BB= =B6=E4=BA=BA: = forwardingalgorithm@ietf.org [mailto:forwardingalgorithm@ietf.org] = =E4=BB=A3=E8=A1= =A8 =E9=9C=8D=E6=9C= =8B=E9=A3=9E
=E5=8F=91=E9=80=81=E6=97=B6=E9=97=B4: 2024=E5=B9=B46=E6=9C=8830=E6=97=A5 12:57
=E6=94=B6=E4=BB=B6=E4=BA=BA: idr@ietf.org; = rtgwg@ietf.org
=E4=B8=BB=E9=A2=98: [Idr] [RTGWG & IDR] = Discussing the Open Scheduled fabric of AI = Network

 

Hi Wg,

 

Link:

 

These documents describe the main disturbances caused = by AI applications to the network and propose solutions based on open = schedulable network architecture developed to address these = issues.

It primarily focuses on the following = aspects:

 

1. Open Scheduled Fabric

  An open network architecture that resolves hash = issues through dynamic scheduling based on network = measurements.

 

2. Bgp Extension for Tunnel Egress Point

  A method for implementing the OSF network = architecture involves using BGP uniformly in the control plane, = extending the data plane as a new tunnel encapsulation, and providing = the necessary control information transmission for the data = plane.

 

3. Extending the BGP protocol to carry Tunnel Egress = Point information within the community attribute.

 

We welcome any comments and discussions.

 

Thank you!

 

Thanks,

Pengfei

= ------=_NextPart_000_0006_01DACECA.83C76750--