[bcause] 答复: PFCP not equal to FMC, Why not choose PFCP

"Songjun (Network)" <song.jun@huawei.com> Mon, 01 April 2019 11:39 UTC

Return-Path: <song.jun@huawei.com>
X-Original-To: bcause@ietfa.amsl.com
Delivered-To: bcause@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 947151200F5 for <bcause@ietfa.amsl.com>; Mon, 1 Apr 2019 04:39:50 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.199
X-Spam-Level:
X-Spam-Status: No, score=-4.199 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id URK73Z-OsO58 for <bcause@ietfa.amsl.com>; Mon, 1 Apr 2019 04:39:47 -0700 (PDT)
Received: from huawei.com (lhrrgout.huawei.com [185.176.76.210]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id AA44712002F for <bcause@ietf.org>; Mon, 1 Apr 2019 04:39:46 -0700 (PDT)
Received: from LHREML713-CAH.china.huawei.com (unknown [172.18.7.107]) by Forcepoint Email with ESMTP id 8B2AEAAB87C994B285F5; Mon, 1 Apr 2019 12:39:44 +0100 (IST)
Received: from DGGEML401-HUB.china.huawei.com (10.3.17.32) by LHREML713-CAH.china.huawei.com (10.201.108.36) with Microsoft SMTP Server (TLS) id 14.3.408.0; Mon, 1 Apr 2019 12:39:42 +0100
Received: from DGGEML509-MBX.china.huawei.com ([169.254.1.28]) by DGGEML401-HUB.china.huawei.com ([fe80::89ed:853e:30a9:2a79%31]) with mapi id 14.03.0415.000; Mon, 1 Apr 2019 19:39:39 +0800
From: "Songjun (Network)" <song.jun@huawei.com>
To: Richard Patterson <richard@helix.net.nz>
CC: "bcause@ietf.org" <bcause@ietf.org>
Thread-Topic: [bcause] PFCP not equal to FMC, Why not choose PFCP
Thread-Index: AQHU5J+7x2DfMtYUgkqHBJV632PtYKYimZnwgAAvkICABGNZ4A==
Date: Mon, 01 Apr 2019 11:39:39 +0000
Message-ID: <F14D3C24A54A484FB585F96F625D2D991E533C05@dggeml509-mbx.china.huawei.com>
References: <BN6PR22MB0771AD2F5A785C1B08318D8B87580@BN6PR22MB0771.namprd22.prod.outlook.com> <F14D3C24A54A484FB585F96F625D2D991E528046@dggeml509-mbx.china.huawei.com> <CAHL_VyBjcomyhKvjW8EB01m3QSifKfonhZCux9vx=OCH30fr+g@mail.gmail.com>
In-Reply-To: <CAHL_VyBjcomyhKvjW8EB01m3QSifKfonhZCux9vx=OCH30fr+g@mail.gmail.com>
Accept-Language: zh-CN, en-US
Content-Language: zh-CN
X-MS-Has-Attach: yes
X-MS-TNEF-Correlator:
x-originating-ip: [10.134.207.212]
Content-Type: multipart/related; boundary="_004_F14D3C24A54A484FB585F96F625D2D991E533C05dggeml509mbxchi_"; type="multipart/alternative"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Archived-At: <https://mailarchive.ietf.org/arch/msg/bcause/fZ6aG5U7G8OsTC-JHbGuqJoaPmM>
Subject: [bcause] 答复: PFCP not equal to FMC, Why not choose PFCP
X-BeenThere: bcause@ietf.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: <bcause.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bcause>, <mailto:bcause-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/bcause/>
List-Post: <mailto:bcause@ietf.org>
List-Help: <mailto:bcause-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bcause>, <mailto:bcause-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 01 Apr 2019 11:39:51 -0000

Hi, Richard

CMCC organized a study on the gaps of PFCP for BNG in the last year, including Nokia, Huawei, ZTE, H3C,etc.  The conclusion is in the email sent by  Qinfengwei in this maillist.

On 2019-03-28, 5:01 PM, "bcause on behalf of qinfengwei" <bcause-bounces@ietf.org<mailto:bcause-bounces@ietf.org> on behalf of qinfengwei@chinamobile.com<mailto:qinfengwei@chinamobile.com>> wrote:
Hi Sanjay and all,
we have analysed the feasibility used PFCP in DBNG for almost half a year with 3GPP team, of course, your colleage Miao has been involved in the discussion.
PFCP is designed specifically for the wireless, there are lots of gaps between wireless and fixed-network, for example PPPoE, L2TP, Route information, detect result , Nat management and so on. Please get analysis result as the following figure, most BNG attributes need be newly defined in PFCP, and as we all know, there is not communication between different manufactures in wireless network, that means there needs a huge amount of work if PFCP is used to support BNG's function. But we have done lots of work on CUSP, and perform lab interoperation test in July 2019. For the massive deployment in 2020, we need the more professional team (such as IETF) to formulate and perfect the standard.
About 82% BNG attributes have to extend or defined.
[cid:Thinkmail@5627590]
Some examples as the following.
CP to send user's flow table to UPs.

Num

Flow Name

Description

3GPP CUSP

1

BRAS_USER_BASIC_INFO

User’s basic information

yes

2

BRAS_USER_PPP_INFO

User’s PPP information

yes

3

BRAS_USER_IFSRV_INFO

User’s access interface

No

4

BRAS_USER_IPV4_INFO

User’s IPv4 address

No

5

BRAS_USER_IPV6_INFO

User’s IPv6 address

No

6

BRAS_USER_QOS_AUTH_INFO

User’s QoS authorization

No

7

BRAS_ROUTEV4_INFO

BRAS IPv4 Routing information

No

8

BRAS_ROUTEV6_INFO

BRAS IPv6 Routing information

No

9

BRAS_STATIC_USER_INFO

Static user information

No

CP to send user's flow table to UPs.


Num

Flow Name

Description

3GPP CUSP


1

BRAS_USER_BASIC_INFO

User’s basic information

yes


2

BRAS_USER_PPP_INFO

User’s PPP information

yes


3

BRAS_USER_IFSRV_INFO

User’s access interface

No


4

BRAS_USER_IPV4_INFO

User’s IPv4 address

No


5

BRAS_USER_IPV6_INFO

User’s IPv6 address

No


6

BRAS_USER_QOS_AUTH_INFO

User’s QoS authorization

No


7

BRAS_ROUTEV4_INFO

BRAS IPv4 Routing information

No


8

BRAS_ROUTEV6_INFO

BRAS IPv6 Routing information

No


9

BRAS_STATIC_USER_INFO

Static user information

No



        IP engineer likes to use IP terms instead of 3GPP’s terms.  This also means changes in PFCP.    200+pages  PFCP standard is not easy to read for IP engineer.
          PFCP uses UDP for high throughput. But in IP network, I do not know which control protocol use UDP.   Using TCP is a simple way to keep control information reliable and in sequence.   CUSP is a lightweight control protocol.

There are many co-operations between BBF and IETF. Normally, BBF defines requirement and architecture, IETF define protocols, such as ANCP.  IETF is more professional in defining protocols.

In any way, I suggest IETF and BBF define protocols by themselves, without lost controlling.    BNG is not a new one. CUSP is just the internal interface of BNG.

Best Regards!

Jun SONG


发件人: Richard Patterson [mailto:richard@helix.net.nz]
发送时间: 2019年3月30日 8:05
收件人: Songjun (Network) <song.jun@huawei.com>
抄送: bcause@ietf.org
主题: Re: [bcause] PFCP not equal to FMC, Why not choose PFCP

Hi Song Jun,

Thanks for your detailed email.

I think a lot of people will have queries around this particular statement of yours regarding PFCP:
•  We found that nearly all requirements of BNG were needed to extended in current protocol. That seems to rewrite a new protocol.

Could you please elaborate on this statement?   Where do you see that PFCP is deficient for fixed line BNG?
If you've got detailed analysis of this, this would be great to see publicly. From what I've heard, PFCP is easily extendable to satisfy the requirements for fixed-line BNG.

You also make the assertion:  •  IETF is the right place to define a protocol for BNG, 3GPP is the right place to extend PFCP.
There is precedent already for the BBF extending protocols such as DHCP and PPPoE with various options.   I think we need to properly understand the extent of how PFCP needs to be extended, in order to decide where best to make the extension, or if we should just build a new protocol.   I do however agree that if it's more than a small extension required, then perhaps the IETF should start from scratch, but I don't see this right now.

-Richard