This if the fourth in a series of guest blogs by Nigel Stephenson, you can find the first three articles here: NFV; the next great use case for shared resource, VMware Sponsors SDN and OpenFlow World Congress, and Choosing the wrong path could halt NFV in its tracks
If nothing’s changed, you aren’t doing it right.
This is a pretty good rule of thumb if you are assessing the potential of your Network Functions Virtualization (NFV) deployment. If you can look at your planned architecture and map the current support model to it, then you should probably take another look. For NFV to deliver the radical cost savings and agility benefits it promises, organisational transformation within operations is mandatory.
This is a certainly a challenge, but it’s not bad news! As in the enterprise world where virtualization is transitioning the role of IT from a cost center to a strategic asset, the CSP operations and support teams will emerge as a key asset for the maintenance and evolution of the service platform and integral to delivering service innovation and cost savings
So what drives the change?
Traditional communication services have been designed, implemented and operated on a project basis. Vertical and monolithic service silos have led to vertical support models with operations teams aligned to specific services.
The move to a shared, virtualized infrastructure is predicated on a service-orientated approach to infrastructure delivery. This creates the need to change how operations teams are organised. The creation, maintenance and ongoing evolution of the infrastructure platform must be abstracted from the higher level VNF services or ‘tenants’ that are deployed upon it. It is this separation of VNF applications from the underlying infrastructure that drives the need for organisational change.
The figure below shows an organisational structure for NFV services modeled on our experience of supporting service oriented IT.
A Cloud Operating Center of Excellence (COE) must be created to bring together the core skills required to support the virtualization platform. The Cloud Operating COE is responsible collectively for the operation, maintenance and ongoing evolution of the virtualized platform.
The Cloud Operating COE is sub-divided into VNF Tenant and NFV Infrastructure operations. The primary technical expertise (administrators, architects etc) resides within the Infrastructure team and act as the technical authority for all aspects of the infrastructure. The tenant operations team acts as the business interface and bridge to other parts of the CSP and to partner organisations. This includes negotiating the service level agreement contract for the VNF, ongoing reporting and (infrastructure) service reviews.
The VNF Tenant Operations team also interfaces to CSP lines of business, development teams and applications vendors to coordinate new service deployment and service innovation.
It is critical to establish such an operational model and define roles and responsibilities in order to facilitate the timely on-boarding of multiple VNFs to the platform. The Cloud Operations COE delivers the control of the platform to the CSP and hence enables the cost efficiencies and service innovation inherent within virtualized deployments.
If you’re looking to manage organisational change, a good place to start is an operational readiness review. Why not ask your VMware representative for more details?