Satoru: Thank you for pointing out further documents.
Responses inline: -----Original Message----- From: ila [mailto:[email protected]] On Behalf Of Satoru Matsushima Sent: Thursday, February 8, 2018 4:00 AM To: Bogineni, Kalyani <[email protected]> Cc: [email protected]; dmm <[email protected]> Subject: Re: [Ila] [E] Re: review comments on ] draft-ietf-dmm-srv6-mobile-uplane-00.txt Hello Kalyani, > [..snip..] > Your slides 9 – 13 show interactions between UPFs and SMF. There are 2 kinds > of UPFs: > Anchor type UPF and service function type UPF. What are the functionalities > of these? Please find some functionalities in the SRv6 mobile Uplane draft: [KB] Your section 5.2 means that UPF closer to radio network is like SGW and your section 5.3 means that UPF closer to Internet is like PGW. When it comes to anchor, it should be equivalent with PSA, PDU Session Anchor, in TS23.501 of 3GPP 5G_ph1 terminology. [KB] in 5G, the functionality of the 2 UPFs can be implemented as one combined function or could be implemented as 2 separate functions with N9 in between. You would also find various SRv6 functions in the network programming draft: Maybe you can see SRv6 mobile uplane as a set of SRv6 functions like a SRv6 profile for mobile with some augment. When it comes to service function type UPF, you name it. Following draft exhibits how service chain can be done by SRv6: [KB] I think these are non-mobility functions as being discussed on the email thread with Marco Liebsch. > What are the changes in SMF functionalities to support SRv6? Is the > interface between SMF and UPFs based on N4/Sx (PFCP in TS 29.244)? SMF functionalities seems still work in progress so that I couldn’t say clearly what the change to it. In CUPS architecture for both Rel-14 and Rel-15, PFCP is expected as Sx and N4 for SRv6 Uplane with no change to over-the-wire messages in basic mode operation: [KB] CUPS does not support branching point function as in 5G. > Also you show IPv6/SRv6 nodes in those slides. Are the UPFs ‘overlaid’ on > IPv6/SRv6 nodes? > Are these UPFs VNFs? Or are UPFs implemented on IPv6/SRv6 nodes? > When you see UPF specifically it should be controlled by SMF through N4, they are not the UPFs. But you might see them as UPFs if a SMF doesn’t control them directly but the SMF can put the sessions to it through some other means. 3GPP SA2 has studied on that case (ETSUN). We consider how SMF deal with that case and SRv6 may help to solve the issues to it in simpler way. Please let me know if you are interested in. [KB] is there a TR for ETSUN that I can read? [KB] I think your document needs to separate out 3 architectures: one for 4G - SGW/PGW; one for 4G - CUPS; and one for 5G - UPF. [KB] I am also still not clear if the blue icons (which I think represent IP/MPLS nodes) in your slides are included in SRv6 architecture or not. Cheers, --satoru _______________________________________________ ila mailing list [email protected] https://urldefense.proofpoint.com/v2/url?u=https-3A__www.ietf.org_mailman_listinfo_ila&d=DwIGaQ&c=udBTRvFvXC5Dhqg7UHpJlPps3mZ3LRxpb6__0PomBTQ&r=IdiSODh8aDRjdCeGgd9MznLHMYKgKcs_YSwXBDiaofh47oilzaXYRYETcBynUdpT&m=dR68JI_amykfLiJQo5cQmTXwKdX_Pn7g-Th7RudMVeM&s=xyt0QGlLtTQfHmezWhutpRYBAw5sEsJlHuzrsihNJbM&e= _______________________________________________ dmm mailing list [email protected] https://www.ietf.org/mailman/listinfo/dmm
