Re: [DMM] DMM WG Adoption Poll (2) for "Mobile User Plane Evolution" - draft-zzhang-dmm-mup-evolution-06

["Jeffrey (Zhaohui) Zhang" <[e-mail]>]

Hi Katsuhiro,

Thanks for your clarification. Please see zzh> below.



Juniper Business Use Only
From: dmm <[e-mail]> On Behalf Of Horiba, Katsuhiro
Sent: Sunday, October 22, 2023 10:07 PM
To: [e-mail]
Subject: Re: [DMM] DMM WG Adoption Poll (2) for "Mobile User Plane Evolution" - draft-zzhang-dmm-mup-evolution-06

[External Email. Be cautious of content]

> Katsuhiro, Jeffrey, please clarify what the control plane means here.
> E.g., (1) control plane in mobility management (MM) and session management (SM) in 5G architecture,
> (2) signaling protocols for N1/2 and N4,
> or (3) control plane for IP routing, BGP, etc.,

In the context I'm referring to, the "Control Plane" encompasses (1) control plane in Mobility Management (MM) and Session Management (SM) in the 3GPP mobile network architecture,
and the act of preserving this information within the (R)AN is what we refer to as "Control Plane Convergence."
Within this context, I believe that the genuine value of the ANUP concept arises from the state in which the (R)AN(-UP) retains MM and SM data through Control Plane Convergence.

Zzh> Looks like in my previous email I misunderstood you 😊 I think we’re on the same page here.

Hence, in alignment with Miya's perspective, we believed that limiting discussions solely to the User Plane would make collaboration with 3GPP quite challenging.

Zzh> I should say that ANUP does not limit the discussion solely to User Plane. As mentioned in the draft and another email, integration of AN and UPF in the user plane allows for convergence/optimization of control plane (now that there is no need to signal the unnecessary N3 tunnel in the targeted use case), and the whole proposal is more valuable with the convergence/optimization of control plane (otherwise it’s just an implementation choice of eliminating the N3 tunnel but still use separate N2/N4 signaling).


However, as David pointed out, it's possible that the IETF Working Group (WG) or the IRTF Research Group (RG) may not be the most suitable forums for addressing this issue.

Zzh> Indeed, eventually the discussion and potential study/work need to happen in 3GPP. This draft is just a starting point for an integrated wireless/wireline architecture, coming from people with IETF/wireline backgrounds who seek DMM as a venue for discussion before it is brought to 3GPP by 3GPP delegates from interested parties.


Lastly, in the context I'm referring to, and as Jeffrey has mentioned, the collaboration with 3GPP's Signaling Protocol, as represented by N1, N2, and N4, is being discussed in draft-mhkk,
and there is no apparent conflict between these discussions and draft-zzhang.

Zzh> Thanks!
Zzh> Jeffrey

Best Regards,
Katsuhiro Horiba