● Networks are vertically constructed by proprietary equipment, causing poor expansibility, great dependence on vendors, and high cost.
● Services, especially highly-customized mobile internet services and internet of things (IoT), are hard to be developed. Traditional closed service development model cannot meet user requirements for rapid customization.
● Proprietary hardware equipment from various vendors exists in the existing network, and network planning, construction, and O&M become increasingly complicated, resulting in high operating costs.
The key to business success is to quickly respond to user demands, shorten time-to-market of new products, and get products out early in the fast-changing market. Therefore, operators have to transform themselves thoroughly in terms of network architecture and operation model. To achieve digital transformation, they need to reconstruct networks by adopting SDN/NFV technologies, focusing on data center, and building an open, agile and intelligent cloud network.
Network reconstruction is generally divided into three phases (Fig.1):
● Infrastructure reconstruction: building a “cloud-network convergence” infrastructure layer.
● Function reconstruction: constructing an “agile and open” service function layer.
● Operation reconstruction: creating a “smart operation” orchestration and management layer.
ZTE’s ElasticNet, a 5G-oriented open network architecture based on the SDN/NFV technology, enables operators to achieve the above reconstructions, provides them with end-to-end telecom cloud solutions, and assists them in network transformation.
Infrastructure Reconstruction: Building a “Cloud-network Convergence” Infrastructure Layer
In a future network, the cloud data center will be as the core, and software will decouple from hardware. With the SDN technology, traffic among various clouds can be optimized, clouds and networks can be deeply converged, and computation, storage, and network resources can be managed and scheduled in a unified way. Networks can be dynamically deployed, and resource pools, such as VMs, bare computers and containers, can be fully used as needed to satisfy user ever-changing demands. In this sense, the new network architecture based on “cloud-network convergence” is the only way to win customers and develop services in the future.
The question is how to make clouds better adapt to telecom applications. ZTE’s TULIP elastic cloud system virtualized data center solution (TECS-VDC) gives the answer. It provides a full range of telecom-enhanced IT hardware products tailored for telecom services, and optimizes NFVI performance with the NUMA, CPU PIN, SR-IOV, and DPDK technologies based on OpenStack optimization. Meanwhile, costs are greatly reduced through hardware acceleration, without affecting the performance of telecom applications. In addition, software-based carrier-class enhancement achieves a system reliability of more than 99.999%, and a reliable disaster-recovery center system implements remote data backup to switch services seamlessly without losing data.
The SDN provides complex connections among clouds, meeting tenants’ requirements for cross-WAN distribution, and achieving cross-domain migration of VMs. It also automatically adapts to service adjustment and implements end-to-end network orchestration through programmable interfaces. Moreover, the SDN-based backbone network can collaboratively optimize the IP layer and optical layer, and adjust services as needed according to the service bandwidth and service level agreement (SLA). This optimizes global traffic scheduling and reduces transmission cost.
Another point of network infrastructure reconstruction is how to use resources in a simpler way. VDC technology realizes one-stop shopping of user resources. With the VDC, users can define the SLA of services, and customize calculation, storage, and network resource specifications. Users can also virtually pool and integrate physical resources, freely create, delete, or change resources, and improve resource utilization through dynamic resource allocation and scheduling. In addition, the automatic service provisioning of VDC can reduce O&M costs and enhance user experience. Now, users can use network-wide resources to innovate services and improve operation efficiency.
Function Reconstruction: Constructing an “Agile and Open” Service Function Layer
In the digital era, users value themselves more than ever and therefore have pressing needs for personalized and customized products and services. Network function reconstruction complies with this trend. Through micro services, componentization, and three core frameworks (development, services, and O&M), network function reconstruction builds a carrier-class innovative NFV PaaS platform for operators to meet the rapid customization requirements of enterprises, individuals, and partners.
With the componentization and network slicing technologies, ZTE’s Network Function Reconstruction enables on-demand generation of networks, experience, and services, which rapidly builds basic network services at low cost. What’s more, the three capabilities (NFV application hosting, efficient Dev development, and Ops optimization) of ZTE’s Network Function Reconstruction create a component-based cloud factory, integrating operation and service management, and attracting numerous developers to cost-effectively introduce personalized services. ZTE’s Cloud Works helps operators achieve network cloudification, business innovation, and management transformation, so as to smoothly evolve to 5G networks.
Operation Reconstruction: Creating a “Smart Operation” Orchestration and Management Layer
ZTE’s MICT-OSTM, as the core of the intelligent management and capacity openness of ElasticNet, is a new-generation management system integrating unified orchestration, automatic O&M, and open operation. By learning from the successful O&M concepts of IT enterprises and adopting SDN/NFV and big-data analysis technologies, ZTE’s MICT-OSTM builds a unified service and resource orchestration and management system.
With this system, services are developed and released in a unified way, and network-wide resources are allocated and extended flexibly. Therefore, the system enables agile service design, automatic operation, and intelligent O&M, thus accelerating innovation and reducing OPEX. With big data analysis and artificial intelligence (AI), ZTE’s MICT-OSTM deeply interacts with consumers, and provides personalized products, services, channels, and contents for users, further improving user experience, creating user value, and achieving business goals.