NFV Plenary #16 wraps up 2016 in Shenzhen
The industry’s principal Network Functions Virtualization (NFV) standards venue wrapped up the year in the world’s powerhouse electronics capital, Shenzhen, December 12-15, hosted by Huawei Technologies. 176 participants under the leadership of newly elected plenary chair Diego Lopez of Telefónica, fielded 80 input documents. The event included multiple NFV sub-group meetings, a joint meeting with the MEC (Mobile Edge Computing) standards group, a Huawei workshop, and a special 5G session. ETSI also provided an update on the first NFV plugtests for interoperability among almost every vendor being held in Madrid for the week beginning January 23.
An additional benefit of the NFV Industry Standards Group is that it sits at the intersection of all other NFV industry groups. So, the plenary meetings provide a one-stop perspective into what is occurring across nearly a dozen other bodies.
The chair provided a useful overview of the unfolding NFV scene across the multiple standards organizations in which he is engaged. He noted that NFV has proceeded from theoretical to rollout at an incredible speed, and that new specification deliverables are being produced rapidly. He noted the NFV Industry Standards Group continues to grow, now up to 300 member organizations, with an incredible 362 meetings for 2016 pursuing 45 active work items.
The focus now is on Release 2 specifications and engaging in close cooperation with 3GPP as the basis for 5G. Notably, ETSI budgeted money for a Specialist Task Force (STF) on IFA/OpenStack gap analysis. Going forward, the principal focus is on 5G NFV implementations for Proof of Concept.
However, significant NFV security and compliance challenges remain. These include not only the introduction of trust mechanisms, but also meeting the array of compliance obligations necessary for public facing infrastructures and industry/government outsourcing of their services. Pervasive encryption is an especially difficult challenge, although the new mcTLS platform could provide a solution at gateways. See http://www.mctls.org
Two final drafts were approved at the plenary – one of which is a significant NFV SEC specification for an “architecture for sensitive components” to ensure isolation of sensitive workloads from non-sensitive workloads sharing a platform, designated GS NFV-SEC 012. A key component is the use of the Trusted Platform Module (TPM) as the basis for hardware root of trust. Yaana Limited (UK) is a supporting organization.
Continuing discussions also occurred concerning network slicing. A particularly useful overview of network slicing treatment across multiple standards bodies was also provided – especially as it relates to 5G instantiations. A key point is that NFV was not redefining Network Slicing, but leveraging NMGN and 3GPP definition – and terminology supporting this definition. A new work item was created on a NFV Network Slicing Technical Report. Network Slices are effected using “blueprints” that provide “a complete description of the structure, configuration and the plans/work flows for how to instantiate and control the Network Slice Instance during its life cycle, and refers to the required physical and logical resources and/or to Sub-network Blueprints.”