Traditionally, telecommunication equipment is provided as physical equipment with software and hardware bound together. Virtualization technologies, in principle, can support network function software to be executed on COTS hardware. The benefits of using virtualization technologies in telecommunication network are:

• Flexibility: Enabling services to be delivered via software on COTS hardware makes the deployment and redeployment of network functions more flexible and allows network functions to be scaled in and out dynamically.
• Time-to-Market: Reducing the time for deployment of new services would improve the time-to-market, thus allowing operators to better support changing business requirements.
• OPEX saving: The decoupling of hardware and software allows the reduction of space, power and cooling requirements and hardware diversity and makes the roll out, management and maintenance of network functions more efficient.
• CAPEX saving: The decoupling of the network function software from purpose-built hardware allows the reuse of COTS hardware for new services and to eliminate wasteful over-provisioning.

Virtualized Networks, like non-virtualized networks, need also to be managed. The present document investigates how virtualized networks can be managed. It investigates the relations of management architecture and management functions between virtualized network, non-virtualized networks and mixed networks. Whether the existing management architecture, management function and management models could be maximum reused and enhanced to satisfy the inclusion of the virtualized network management function are also studied.

The present document studies the network management of virtualized 3GPP specified core networks based on the architectural framework in ETSI GS NFV-MAN 001 [2]. It studies the network management of mobile networks that include virtualized core networks functions. It also identifies the potential impacts on the existing 3GPP management architecture and to propose potential solutions for the network management of fully virtualized networks and mixed networks. The objectives of this study are:

• Study the use cases and concepts for the network management of virtualized networks, which are applicable to 3GPP, taking into account the relevant use cases from ETSI.
• Analyse and classify the network management scenarios when all instances of 3GPP-defined network elements in a subsystem/domain are virtualized according to the architectural framework in ETSI GS NFV-MAN 001 [2] (i.e. in fully virtualized networks).
• Identify the requirements for potential solutions for the network management of virtualized networks when all instances of 3GPP-defined network elements in a subsystem/domain are virtualized according to the architectural framework in ETSI GS NFV-MAN 001 [2] (i.e. in fully virtualized networks).
• Analyse and classify the network management scenarios when some instances of 3GPP-defined network elements in a subsystem/domain are virtualized according to the architectural framework in ETSI GS NFV-MAN 001 [2] (i.e. in mixed networks).
• Identify the requirements for potential solutions for the network management of virtualized networks when some instances of 3GPP-defined network elements in a subsystem/domain are virtualized according to the architectural framework in ETSI GS NFV-MAN 001 [2] (i.e. in mixed networks). Study whether or not a single management system for mixed networks is required.
• Identify the potential impacts on the existing 3GPP Management reference model.
• Analyse the existing 3GPP Management reference model, interfaces, protocols and procedures to determine what can be re-used, adapted or extended and whether new IRPs are needed for network management of virtualized networks.
• Propose enhancements or extensions to the 3GPP Management reference model, if an impact is foreseen.
• If needed, solution(s) for a single management system for mixed networks should be studied.
• Provide recommendations for the standardization of the network management of fully virtualized networks and mixed networks, based on the result of the analysis and the potentially identified impacts, enhancements or extensions.