You're correct. The separation work mentioned below is SDN focused but multiple protocols between the Controller, acting as a FPC Agent, and DPN are used.
-----Original Message----- From: Behcet Sarikaya [mailto:sarikaya2...@gmail.com] Sent: Monday, August 29, 2016 10:15 AM To: Bertz, Lyle T [CTO] <lyle.t.be...@sprint.com> Cc: dmm@ietf.org; lionel.mor...@orange.com Subject: Re: [DMM] DMM FPC I-D - Implementation As I mentioned in my previous mail, 3GPP already supports control plane data plane separation, they just call it user plane control planes. See: https://en.wikipedia.org/wiki/System_Architecture_Evolution I think you are confused by the SDN approaches in the control plane in 3GPP. However, 3GPP decided not to pursue SDN. But they will pursue NFV instead. FYI. Regards, Behcet On Fri, Aug 26, 2016 at 3:00 PM, Bertz, Lyle T [CTO] <lyle.t.be...@sprint.com> wrote: > All, > > > > Last week at the Intel Developer Forum in San Francisco (US), Intel > and Sprint demonstrated a virtual Evolved Packet Core (vEPC) that > included, amongst other features, a SGW and PGW (3GPP functions similar to > MAG and > LMA). This software > > 1. Used a Separated Control and Dataplane Node with an intermediate > SDN Controller – Opendaylight’s Beryllium release. > > 2. Used IETF DMM FPC-v03 > (https://datatracker.ietf.org/doc/draft-ietf-dmm-fpc-cpdp/?include_tex > t=1) for communication of some sessions (others were simulated to > ensure we met the demonstration timelines). > > > > The Control and Dataplane software performs well (10Gbps down / 3Gbps > up for 250K users) on a single Intel processor’s *core* based upon the > open source DPDK libraries from Intel. The demonstration shows the > promise of separation of Control and Data planes. > > > > We had to deviate quite a bit from the version 03 specification to > support 3GPP but version 04 adds the 3GPP model as an option. Many of > the general changes in version 04 are influenced by the implementation. > > > > Work is already underway to implement version 04 with several internal > changes, e.g. moving from the Opendaylight MD-SAL (model driven) > approach to the application driven (AD-SAL) approach, as focus is now > on performance. I hope to share more details in the future. As a > part of the final version of the FPC I-D we will submit an > Implementation Status section in the draft per RFC 7942. > > > > Lyle Bertz > > Sprint > > > ________________________________ > > This e-mail may contain Sprint proprietary information intended for > the sole use of the recipient(s). Any use by others is prohibited. If > you are not the intended recipient, please contact the sender and > delete all copies of the message. > > _______________________________________________ > dmm mailing list > dmm@ietf.org > https://www.ietf.org/mailman/listinfo/dmm > ________________________________ This e-mail may contain Sprint proprietary information intended for the sole use of the recipient(s). Any use by others is prohibited. If you are not the intended recipient, please contact the sender and delete all copies of the message. _______________________________________________ dmm mailing list dmm@ietf.org https://www.ietf.org/mailman/listinfo/dmm
