Knowledge dissemination of 5G network slicing

Knowledge dissemination of 5G network slicing 5G network slices can be divided into multiple logically independent virtual networks on the same physical network infrastructure. Each network slice is an isolated end-to-end network with its own unique latency, throughput, security and bandwidth characteristics, allowing flexibility to address different needs and services. The one-size-fits-all network model used in the past for mobile networks (2G, 3G and 4G) no longer fits today's changing market model, where each use case has its own unique performance requirements, making the one-size-fits-all approach to service delivery obsolete. The future network needs to transition from "one size fits all" to "one size per service" through network slicing technology. Network slicing is a key technology in 5G networks that promises to transform "best effort" networks into networks that provide higher reliability, using a single physical network infrastructure to accommodate different and varying quality of service (QoS) requirements. 5G network slices can be divided into multiple logically independent virtual networks on the same physical network infrastructure. Each network slice is an isolated end-to-end network with its own unique latency, throughput, security and bandwidth characteristics, allowing flexibility to address different needs and services. Network Slicing Architecture In the abstract, the network slicing architecture consists of two main modules, one for the actual slicing implementation and the other for slicing management and configuration. The first module is a multi-layer architecture consisting of a business layer, a network function layer, and an infrastructure layer, each performing a different task for slice definition and deployment. The second module is designed as a centralized network entity, often referred to as a network slice controller, that monitors and manages the functions among the three layers in order to efficiently coordinate the coexistence of multiple slices. Business Layer The business layer interfaces directly with network business entities that share the underlying physical network (e.g., MVNOs and third-party service providers) and provides a unified view of services on demand. Each service is represented as a service instance, embedded in the network characteristics according to SLAs (Service Level Agreements), with the expectation that the service requirements are fully satisfied by creating the appropriate slices. Network Functionality Layer The Network Functionality Layer is responsible for creating network slices based on service instance requests from the upper layers. It consists of a set of network features that contain well-defined behaviors and interfaces. Multiple network functions are placed on the virtual network infrastructure and connected together to create an end-to-end network slice instance that reflects the network characteristics of the service request. To increase resource utilization, different slices can share the same network functions at the same time, but this increases the complexity of operational management. A one-to-one mapping of network functions to slices simplifies the configuration process, but results in low resource utilization. Infrastructure Layer The infrastructure layer represents the actual physical network topology (wireless access network, transport network, and core network) that can be reused by each network slice and provides the physical network resources to host the multiple network functions that make up each slice. The available physical network resources include a heterogeneous set of infrastructure components such as data centers (storage and computing power resources), devices supporting network connectivity such as routers (network resources) and base stations. Network Slicer Controller A network slice controller is defined as a network coordinator that interfaces with the various functions performed at each layer to manage each slice request in a uniform manner. The benefit is that it allows efficient and flexible slice creation and reconfiguration during the slice lifecycle. Operationally, the Network Slice Controller is responsible for the following tasks, providing effective orchestration between the three tiers. End-to-end service management: Mapping the various service instances represented in the business layer according to the SLA requirements to the appropriate network functions in the network function layer that conform to the service instance constraints. Virtual Resource Definition: Virtualization of physical network resources to simplify resource management operations when assigning network functions. Slice Lifecycle Management: Slice performance monitoring across all three layers to dynamically reconfigure each slice to accommodate possible SLA requirement modifications. Due to the varying complexity of each task, the network slice controller can consist of multiple schedulers that independently manage a subset of functions at each layer. To meet the service requirements, the various scheduling entities need to coordinate with each other by exchanging high-level information about the operational states involved in slice creation and deployment. As we mentioned earlier, network slices are virtual networks segmented on the same physical network infrastructure. Since they share the same infrastructure, how can we ensure that each slice is independent of each other and that the failure of one will not affect the other slices? Slice Isolation Slice isolation is an important requirement, that is, multiple slices sharing the same infrastructure coexist at the same time, while ensuring that the performance of each slice cannot affect the performance of other slices. Slice security: A network attack or failure only affects the target slice and has limited impact on the lifecycle of other existing slices. Slice privacy: The private information associated with each slice (e.g., user statistics, MVNO business model) is not shared with other slices. Before slicing, the first thing to consider is how to unify the management of each module after slicing to form an organic whole? To do this, we have to start from SDN and NFV technology. SDN and NFV Network slicing uses SDN (Software Defined Networking) and NFV (Network Function Virtualization) to quickly segment the network and its resources to support specific applications and devices. SDN separates the network control plane from the data plane and manages network traffic through application programming interfaces (APIs) in the central control plane. The control plane allocates resources through the business layer to provide customized services to the client. NFV is another prerequisite for slicing, and the strategy behind NFV is to install network functions onto virtual machines (VMs) on virtual servers to provide services that run on traditional proprietary hardware. SDN is used to control the configuration of VMs located at the edge or in the core cloud, and NFV can also be used to manage the lifecycle of network slices and their infrastructure resources. These technologies work together to provide a good foundation for SDN and NFV networks to effectively leverage virtual and physical resources, including RANs. Relying on SDN/NFV technology, we can abstract all hardware into three types of resources, namely computing, storage and network, for unified management and allocation, and give different resources to different slices, and completely isolate them from each other, realizing logical high-level unified management and flexible cutting. 5G Private Network In addition to network slicing, there is another way to meet different business needs - 5G private network. A private 5G network is a local area network (LAN) that uses 5G technology to create a dedicated network with unified connectivity, optimized services, and secure communications in a specific area. A private network is a dedicated network that provides network communication services to specific users. The difference between a public network and a private network is that a public network serves the general public, while a private network serves a specific audience. Enterprises seem to prefer dedicated 5G networks to network slicing. 5G private networks offer enterprises the freedom to customize their networks, providing different configurations based on the location and type of work, with significant privacy and security advantages. Private networks typically have their own devices and device authentication and licensing," said Don Alusha, senior analyst at ABI Research. 5G network slices tailor network services to different granular demand ranges, and reliability is largely guaranteed by the operator," said Don Alusha, senior analyst at ABI Research. Why does a company need network slicing when it can create its own private network? Security, privacy and trust are key requirements for every business application. While 5G is one of the more secure cellular generations, touting advanced encryption capabilities, it is still vulnerable to attacks. As a result, many enterprises are wary of sharing network slices across carriers' public 5G networks, preferring to control their own private networks. "Most enterprises want to control their own closed systems and keep them completely separate from the public infrastructure," Antlitz said. "Another key reason is trust. They don't trust that there won't be vulnerabilities or other problems with slices of the network. So will 5G private networks really kill network slicing? Enterprises need to consider deployment options when evaluating 5G. One option is to build and control a dedicated 5G network using their own spectrum. The second option is to take network slices from the carrier's public 5G network. This choice depends heavily on trust, technology maturity, and cost considerations. Building a private network is not an easy task, and the first thing to face is spectrum acquisition. In most countries, spectrum is considered a natural resource and its use is controlled by the national authority, which allocates resources according to the country's needs. Second, the cost of deploying a dedicated 5G network is high, and ordinary enterprises, especially small enterprises, cannot afford to purchase and deploy a full 5G network, and will need some professional operation and maintenance staff to maintain the network in the later stage. By using the carrier's public network, users can get 5G services supported by SLAs, while avoiding the hassle of buying spectrum and maintaining the network. So, while enterprises are wary of privacy and security concerns, network slicing will remain an attractive option for SMBs. Total Potential Market for 5G B2B Applications, Global Forecast Period 2020-2030 (Source: ABI Research) Challenges and Opportunities Challenges Despite the tremendous benefits of network slicing, operators and developers still face many challenges. Complete end-to-end network slicing spans the wireless access, transport, and core networks, but these RANs need to be redesigned to accommodate network slicing. While the standardization process continues, there is still a vague industry-wide consensus on the deployment of network slicing for 5G networks and other architecture elements. Adding more networks to the same physical infrastructure puts additional pressure on operators to maintain SLAs, QoS and security guarantees for each individual slice. Opportunity The opportunity is self-evident. With the explosion of multiple business and application scenarios such as car networking, industrial Internet, VR/AR, etc., network slicing can divide a single network into "VIP networks" that meet different SLAs for different use cases based on customer demand. Offering network slices as a service minimizes operational expenses (OPEX) and capital expenditures (CAPEX). For critical public applications, such as first responders and medical emergency teams, network slicing gives priority in terms of coverage, capacity and connectivity. Network slicing will change our lives and our production.