Network slicing overlays multiple virtual networks on top of a shared network, in which each slice of the network can have its own logical topology, security rules, and performance characteristics, within the limits imposed by the underlying physical networks. Each network slice is an isolated end-to-end network tailored to fulfill the requirements requested by a particular application. This can act as an efficient solution to address the diverse requirements of a mobile network, especially in the case of a 5G network, and provide the flexibility and scalability associated with future network implementations.
For network slicing to occur, automation is considered to be an important component, as it is expected that mobile network operators will have to design and maintain hundreds or thousands of network slices. The mobile network operators cannot manage this volume of slices manually at the speeds expected to be required by its customers. Instead, end-to-end automation is expected to be used to perform zero-touch slice lifecycle management at scale and in real-time, as traffic load, service requirements, and network resources change.
Slicing technologies on ethernet networks are as old as virtual local area networks (VLANs). The concept has been more fully realized with the rise of software-defined networking and software-defined wide area networking. SDN separates the network's control plane from the packet handling data plane and enables the control plane to define virtual networks by defining packet handling rules and pushing those rules out to the data plane devices.
In effect, network slicing allows for the slicing of a single physical network into multiple isolated logical networks, which can offer a plethora of services with different service level requirements. This can, in turn, help towards a realization of a service-oriented view of the network using concepts from software-defined networking and network function virtualization. From a business perspective, this would allow each network slice to be administered by a mobile virtual network operator (MVNO). By this, an infrastructure provider can lease its physical resources to MVNOs, and the MVNO can deploy multiple network slices over the physical network customized to the applications needed by its users. Network slicing can offer new business opportunities for communications service providers and a range of use cases and sectors by making it possible to create fit-for-purpose virtual networks with varying degrees of independence.
In this virtualization scenario, the physical components are secondary, while logical (or software-based) partitions become paramount, devoting capacity to certain purposes dynamically and according to need. And, as those needs change, the resources can change. Using common resources such as storage and processors, network slicing permits the creation of slices devoted to logical, self-contained, and partitioned network functions.
Network slicing can act as an efficient solution to address the diverse requirements of 5G mobile networks and offer a framework for network implementations. It allows 5G network architecture and those operators to provide portions of their networks for specific customer use cases, whether the use case is the smart home, the IoT factory, connected car, or the smart energy grid.
A 5G network slicing example.
Network slicing is considered to be overall an important technology for 5G as new services will need different requirements, which will require different throughput, latency, and reliability. The use cases for 5G and network slicing could fall into three broad categories:
- Extreme (or enhanced) mobile broadband (eMBB): these applications are video-centric and consume a lot of bandwidth and will generate the most traffic on the mobile network
- Massive machine-type communication (mMTC): this is also known as the Internet of Things, but at a much larger scale, with billions of devices being connected to the network. These devices will generate far less traffic than eMBB applications
- Ultra-reliable Low-Latency Communications (urLLC): these would allow for things like remote surgery or vehicle-to-x (v2x) communications and require mobile network operators to have mobile edge computing capacity in place
There is a concern that private wireless networks may stop network slicing before it can be properly deployed. In an instance of a private wireless network, a company can create its own private network that has its own authentication and authorization for devices and equipment. Whereas slicing customizes network services according to a different range of granular requirements and reliability is guaranteed by service providers. Private networks are considered a more static deployment, while network slicing is more associated with dynamic deployments. While network slicing is based on a carrier's full network, a private wireless network is often more localized and an enterprise can fully control a private wireless network through an IT department and provide secure connectivity within a prescribed area.
However, private wireless networks tend to be high-cost solutions, especially as the need for the network to be extended can increase the overall cost of the network. Whereas slicing is relatively low-cost, because vertical industries use operator deployed network assets. The overall cost of either deployment also depends on the size of the enterprise, the spectrum, and the data requirements. Many companies developing the necessary technology for network slicing are also developing the necessary technologies for private wireless networks.
