Hi Kalyani, I was thinking something along this:
A section on each competing architecture where we first provide a description of each architecture. A section on the list and set of features that 3GPP requires and must be supported. In the following sections we then make a deep dive into each architecture and describe/analyse how different architectures would support each of the requested features. Going through the above exercise we can perhaps tabulate the outcome of the investigation in a format like: Arch1 Arch-2 Arch-N Feature-1 Feature-2 Feature-N Considering pros and cons of each architecture against the request features, at this point we should be ready to reply back to 3GPP with our recommended architecture. We then continue the work, and go into more details and follow the same format for different protocols supporting the recommended architecture and tabulate the outcome as Arch1.1 Arch1.2 Arch1.n Feature-1 Feature-2 Feature-N Taking ID-Location architecture as an example, LISP/ILSR, ILNP, ILA, etc. will be Arch1.1, Arch1.2, Arch1.3, etc. Now, here is my question. Have we already decided to go with the recommendation that ID-Location architecture should be the next generation of data path? If everyone is on board with this, then your suggested outline fits the bill and we can start working on it right away. If the decision hasn't been made then it would be perhaps a good idea to go one level up and talk about different architectures first. Mind you that I don't know and am simply curious to know what our starting point is. Arashmid From: Bogineni, Kalyani [mailto:[email protected]] Sent: 18 January 2018 10:21 To: Arashmid Akhavain <[email protected]>; Sri Gundavelli (sgundave) <[email protected]>; [email protected] Cc: AshwoodsmithPeter <[email protected]>; TongWen <[email protected]>; Bogineni, Kalyani <[email protected]> Subject: RE: [DMM] white paper for optimized mobile user plane solutions for 5G Arashmid: The individual sections are supposed to allow the proponents of the protocols to show how the 5G architecture will be when their protocol is used with the focus on N9. The criteria section lists the evaluation criteria from architectural perspective. Do you see a need for a different kind of format? Kalyani From: Arashmid Akhavain [mailto:[email protected]] Sent: Thursday, January 18, 2018 9:38 AM To: Sri Gundavelli (sgundave) <[email protected]<mailto:[email protected]>>; Bogineni, Kalyani <[email protected]<mailto:[email protected]>>; [email protected]<mailto:[email protected]> Cc: AshwoodsmithPeter <[email protected]<mailto:[email protected]>>; TongWen <[email protected]<mailto:[email protected]>> Subject: [E] RE: [DMM] white paper for optimized mobile user plane solutions for 5G Hi Sri, Thank you for your reply and clarifications. I think Kalyani's work is extremely valuable, but we also need to set an strategy for our reply back to 3GPP. Currently most discussions are about individual protocols such as SRV6, LISP/ILSP, ILNP, ILA, etc. which are really in one form or another a type of ID-Location approach. I believe we first need to address 3GPP's request from the architecture point of view, before we dive into implementation details. Is there any discussion around different competing network architectures? Not sure if you have seen a white paper titled "Evolving 5G Routing" by Docomo, Huawei and others. We also hosted a live webcast demo that shows two LTE slices using ID-Location for data path and distribute ID-Location information to interested eNodeBs via public cloud pub/sub service. The result is very promising and shows the capability of this architecture. I believe the combination of the white paper and the demo could perhaps serve as a good starting point to get the architecture discussion going and that's where conference calls can come handy. I agree though let's start with the mailing list and move to calls next. Best regards, Arashmid Akhavain From: Sri Gundavelli (sgundave) [mailto:[email protected]] Sent: 17 January 2018 17:53 To: Arashmid Akhavain <[email protected]<mailto:[email protected]>>; Bogineni, Kalyani <[email protected]<mailto:[email protected]>>; [email protected]<mailto:[email protected]> Subject: Re: [DMM] white paper for optimized mobile user plane solutions for 5G Arashmid - We do have regular DMM working group meeting during the IETF week. I guess, what you are asking for is a WG-chair scheduled conference calls? DMM List may be very quite, but there have been regular conf calls between authors of some of the WG documents. Authors of FPC host those calls regularly. For the new work item that Kalyani is proposing, we need to have some discussions in the mailer and once the document is adopted as a WG document, we can certainly host some calls (author/chair scheduled) on a need basis. So, the first step is to post comments on the documents and trigger some discussions. Sri From: dmm <[email protected]<mailto:[email protected]>> on behalf of Arashmid Akhavain <[email protected]<mailto:[email protected]>> Date: Wednesday, January 17, 2018 at 2:16 PM To: "Bogineni, Kalyani" <[email protected]<mailto:[email protected]>>, "[email protected]<mailto:[email protected]>" <[email protected]<mailto:[email protected]>> Subject: Re: [DMM] white paper for optimized mobile user plane solutions for 5G Hi everyone, Online meetings on a regular basis can perhaps help moving things forward. Arashmid From: dmm [mailto:[email protected]] On Behalf Of Bogineni, Kalyani Sent: 17 January 2018 15:20 To: [email protected]<mailto:[email protected]> Subject: Re: [DMM] white paper for optimized mobile user plane solutions for 5G Folks: Here is the draft with 3GPP architecture described as much as relevant to N9 interface. I have extracted some diagrams from 3GPP specs and will redo them in ASCII after I get some feedback on whether to include them or not in this document. Kalyani From: Bogineni, Kalyani Sent: Wednesday, December 13, 2017 4:14 PM To: [email protected]<mailto:[email protected]> Cc: Bogineni, Kalyani <[email protected]<mailto:[email protected]>> Subject: white paper for optimized mobile user plane solutions for 5G Folks: In response to the 3GPP CT4 study item on user plane protocols, we propose that a white paper be developed that can compare the different IETF protocols. Attached is an outline. Please let me know if you would like to contribute. Kalyani Bogineni Verizon
_______________________________________________ dmm mailing list [email protected] https://www.ietf.org/mailman/listinfo/dmm
