Hi, Thanks for the clarification below. I think the document could be a bit clearer on the intention of the different aspects of the document. For example the Socket API parts being declared more clearer as not being a specification and only an exploration of the potential solutions. Regarding TAPS, that comment was truly about potential for further improvements to both DMM's and TAPS goals by cooperation. TAPS impact on this document could at most be an informational statement in section 3.4 (Conveying the Desired Address Type) but also that may be premature as the mobility related aspects expands the TAPS scope somewhat to my understanding. Cheers Magnus On 2019-01-09 12:38, Moses, Danny wrote: > Thanks for the feedback. I think I understand the concern. Perhaps it is partly due to not participating in the presentations and discussions that took place in DMM over the past several years. > > Let me try and add some more context. > The desire to optimize mobility services is part of DMM's charter. By the way, similar work was also carried out the 3GPP (the SSC feature - Service and Session Continuity) for Release-15 - the first 5G release and we have tried to stay in sync with that work. > > Handling such optimization requires the following steps: > 1. Having the application express its session continuity preferences. > 2. Conveying these requirements to the mobile network. > 3. Selecting the session continuity scheme by the network. > 4. Notification: Network to mobile node. > 5. Providing results to the application > > We have developed several drafts that deals with entire flow. This draft defines the different service levels and handles steps (1) and (5). There is another individual draft - draft-moses-dmm-dhcp-ondemand-mobility-10 that handles (2) and (4) via DHCPv6 extension and draft-feng-dmm-ra-prefixtype-03 that handles (4) via an RA (Router Advertisement) option. > > There were several discussions as to whether this draft should specify Socket extensions or provide guidelines for an API provided by the network stack to applications. The decision, eventually, was that since IETF does not specify the Socket API, we should not specify Socket extensions, but rather, provide guidelines for such functionality. > > As for TAPS I can prepare a topic to be discussed in the WG to see if there is any interest in this work over there. I hope however, that this is not gating the approval of this draft. > > Can we discuss TAPS separately? > > > > -----Original Message----- > From: Magnus Westerlund [mailto:magnus.westerlund@xxxxxxxxxxxx] > Sent: Tuesday, January 08, 2019 18:59 > To: tsv-art@xxxxxxxx > Cc: draft-ietf-dmm-ondemand-mobility.all@xxxxxxxx; ietf@xxxxxxxx; dmm@xxxxxxxx > Subject: Tsvart last call review of draft-ietf-dmm-ondemand-mobility-15 > > Reviewer: Magnus Westerlund > Review result: Ready with Nits > > This document has been reviewed as part of the transport area review team's ongoing effort to review key IETF documents. These comments were written primarily for the transport area directors, but are copied to the document's authors and WG to allow them to address any issues raised and also to the IETF discussion list for information. > > When done at the time of IETF Last Call, the authors should consider this review as part of the last-call comments they receive. Please always CC tsv-art@xxxxxxxx if you reply to or forward this review. > > First of all I do become a bit uncertain about the intentions of this document. > As an informational document I think discussing an possible optimization and how it can be solved is all okay. What I fail to see the point and a likely a source of confusion is the draft socket API changes which may be considered as solutions. However, an detailed solution to the problem space requires one to actually dig into some of the areas the document explicitly calls outside of its intentions. Thus, I wished the document was a bit clearer on its purpose of only sketching an idea and be firmer of not actually offering a ready solution that can be implemented. Thus, I think there are risks with having something that appears to define a socket API extension. If the intention is to actually define socket API extensions then I think there are much more that needs to be defined and solved. > > Secondly, I think the proponents of this work should have a long and serious discussion if the ongoing work in the TAPS WG can actually provide an better way forward for the API as well as provide an improvement to the TAPS architecture. Because if an application specifies its needs for session continuity then an TAPS implementation could fulfill this either using a combination of TCP with Session lasting IP address or with Non-persistent IP address and transport protocols that has built in session mobility or continuity features such as MPTCP or QUIC. > > > --------------------------------------------------------------------- > A member of the Intel Corporation group of companies > > This e-mail and any attachments may contain confidential material for > the sole use of the intended recipient(s). Any review or distribution > by others is strictly prohibited. If you are not the intended > recipient, please contact the sender and delete all copies. > -- Magnus Westerlund ---------------------------------------------------------------------- Network Architecture & Protocols, Ericsson Research ---------------------------------------------------------------------- Ericsson AB | Phone +46 10 7148287 Torshamnsgatan 23 | Mobile +46 73 0949079 SE-164 80 Stockholm, Sweden | mailto: magnus.westerlund@xxxxxxxxxxxx ----------------------------------------------------------------------
