Summary Bullets:
• This week’s SDN World Congress looks to explore the necessary connections between SDN, NFV and future 5G networks; with past experience and the event agenda as a guide, don’t expect any real revelations or insights.
• What the industry needs to begin grappling with are the specific SDN solution requirements driven by 5G along with deeper details on how fundamental technologies like NFV Network Slicing will operate.
This week, Layer 123’s SDN & OpenFlow World Congress takes place in The Hague. Since its inception, the event has been one of the most important gatherings for understanding the state of SDN (as well as NFV) and how the technology is evolving. Vendors show up to talk about their new products. More importantly, service providers show up to talk about their progress with SDN and what they’re looking for from the market. This is why we ran a webinar last week to discuss our expectations out of the show.
Given the appearance of a “5G Core and Transport Forum” on the Day One and Day Two agendas, it’s clear that one thing the organizers hope to do is drive a conversation around the intersection of SDN with 5G. With the Radio Access Network (RAN) dominating most 5G discussions, it’s nice to see the core network get some much needed attention; after all, you can’t build a 5G network from base stations alone. That said, I’m not sure anyone should get too optimistic or expect grand revelations about the 5G core coming out of The Netherlands by week’s end.
To date, most of the 5G core discussions I’ve taken part in at trade shows have followed a similar pattern, confining themselves to a handful of similar conclusions: the 5G core will be built on a foundation of virtualized network functions; supported by SDN, network slicing to allow for support of multiple, varied use cases; thank you very much for joining us. This is all fair. But it’s not particularly insightful around the details of how a 5G core will leverage SDN or NFV, how slicing will operate in practice, and what requirements 5G will place on those technologies. And I don’t expect this year’s SDN and OpenFlow World Congress doing much to remedy this lack of insight. In part, standards around 5G core architectures are still getting built out. Beyond that, the event’s agenda just isn’t that encouraging. Combining transport and core into one track necessarily limits the focus that either gets. That’s a given. Beyond that, session topics that consider “NFV/SDN as a key technology enabler,” “evolution and role of packet microwave,” or “impacts of NFV on testing mobility” don’t inspire confidence that we’ll come away with a deeper understanding of how 5G core networks will be built.
So, what am I looking for? What conversations do I think we should be having?
To the extent that the 5G core won’t be built from generalities, we need to begin talking about the direct linkages between 5G use cases and SDN/NFV requirements. What will SDN and NFV solutions need to do in order to accommodate 5G? What do specific 5G use cases (e.g., massive IoT, critical communications, and enhanced mobile broadband) imply for SDN and NFV solution capabilities? And, if one of those capabilities is support for Network Slicing, how will this operate in practice – not just theory – in the context of an increasing push towards edge-based computing?
With presentations from Vodafone, Huawei, EE, NTT DOCOMO, and Tele2, I’m holding out hope that we will hear how 5G theory is beginning to connect with SDN and NFV solution sourcing and planning. If we want to build actual (not theoretical) 5G networks, we need to begin having those discussions soon.