The recent progress in communications and sensors have paved the path for the ever-growing development of Internet of Things (IoT) services, where a tremendous number of devices demand access to the transport network, using commonly deployed fixed or wireless access technologies or even mobile Radio Access Network (RAN).
Supporting IoT in the RAN is challenging, as IoT services may generate a multitude of short and bursty sessions, thus impacting the performances of mobile users sharing the same RAN.
To this end, Network Slicing is envisioned as a promising design approach to enable optimal support for heterogeneous service segments sharing the same RAN, i.e., the key requirements of the upcoming fifth generation (5G) mobile network.
The fifth generation (5G) network is expected to accommodate a multitude of services, with heterogeneous requirements. The International Telecommunication Union (ITU)and Fifth Generation Public-Private Partnership (5G-PPP) have identified three categories of services to be handled by 5G :
- Enhanced Mobile Broadband (eMBB)
- massiveMachine-Type communications (MMTC)
- Ultra-Reliable and Low Latency communications (URLLC)
Among these categories, several use-cases can be defined, varying from general broadband cellular systems to Internet of Things (IoT) networks. All these services require a set of heterogeneous requirements that cannot be satisfied by the traditional one-size-fits-all architecture.
Therefore, alternative architectural choices are required to handle all these diversified services in an efficient manner.
With this in mind, 5G is being designed with a programmable and a flexible infrastructure, allowing different services (e.g., IoT, cellular, vehicular, etc.) to share the same radio access network (RAN), while guaranteeing Quality of Service (QoS) and Service Level Agreement (SLA).
Besides, the concept of Network Slicing is employed for enabling the efficient coexistence of heterogeneous services in the same shared network.
What Is Network Slicing?
Network slicing is the operators’ best solution on how to construct and manage a network that connects and exceeds the emerging requirements from a wide range of enterprises.
The way to innovate a sliced network is to modify it into a set of logical networks on top of a shared infrastructure. Each logical network is implemented to serve a defined business purpose and comprises all the required network resources, configured, and connected end-to-end.
Network Slicing In Emerging Network Virtualization Technologies
According to the Next Generation Mobile Networks (NGMN) Alliance’s vision for 5G, slices are defined as end-to-end self-contained logical networks, which can be controlled and managed in an independent way by the slices’ owners, such as Over-The-Top (OTT) service providers and Mobile Virtual Network Operators (MVNOs).
The new emerging network virtualization technologies, like as Software Defined Networking (SDN) and Network Function Virtualization (NFV), are envisioned as the key enabling approaches of network slicing, by providing the required tools to virtualize the physical resources and allocate them to the logical slices in an efficient way.
Moreover, SDN and NFV provide the appropriate tools to orchestrate the underlying resources and enables the coexistence of independent slices within the same physical infrastructure.
Introduction To Network Slicing In IoT Service Deployment
The potential number of connected devices, the massive amount of data these devices generate, and the growing complexity of the Internet of Things ((IoT) infrastructure, is a high challenge to build future IoT applications.
Such applications exploit a set of characteristics, such as heterogeneity, interoperability, dynamicity, and geographical distribution. Additionally, the quality of service (QoS) notion is particularly relevant in this context, because it affects user experience, resource consumption, and energy efficiency, among other things.
To cope with the QoS problem, solutions have to be defined not only at the network level but also within the IoT platforms on which the applications are distributed.
Background And Key Issues For Network Slicing In IoT Service Deployment
- The IoT applications QoS requirements
The IoT is based on the connection of billions of objects, going beyond laptops and smartphones, including connected cars, wearables, smart cities, smart homes, among others. Under the current circumstances of COVID-19, innovative IoT based applications are going to emerge in diverse domains such as smart health, telemonitoring, thermal scanning, face recognition, contactless payments, smart wearables, transport etc.
Those future applications will have specific QoS needs (bounded response time, availability, etc.) that will have to be considered by the IoT platforms such as the one promoted by the One M2M IoT platform QoS bottleneck.
Such platforms are formed by heterogeneous infrastructure nodes, typically servers and gateways having different resources/capabilities.
These nodes represent potential bottlenecks in network slicing with respect to the QoS, for instance when a too high number of application-level requests have to be processed by a given node.
- Existing solutions to tackle QoS in IoT limitations
Several solutions have also been proposed that address the QoS issue at the platform level. Most of them are based on a service differentiation principle that allows processing the requests differently, depending on their priority.
Here, the services managed on each node are the ones that have been provided at the initialization of the platform, such as traffic marker/shaper, message or task scheduler, etc.
This principle is adequate when the available services allows tackling all the application requirements. However, it becomes inadequate when a service is not existing on a node, or when the computing resources are not sufficient. The network slicing is a new concept that allows answering this limitation.
- Network slicing concept
The maturity of technologies such as SDN and NFV, allows to consider new network concepts such as slicing networks that will allow a more flexible management of networks. The ITU-T defines a network slice as a logical network that provides specific network capabilities and network characteristics.
The (network) slicing consists in building slices on demand. It has been initially thought to share resources on a communication infrastructure. It is now more and more considered to perform QoS provisioning.
The slicing concept is based on the notions of network function (NF) and NF Service chaining. Basically, a NF is a processing function applied on a given data traffic (e.g. a delaying function applied on IP packets).
More formally, a NF is defined by 3GPP as a processing function in a network, which has defined functional behavior and interface.
NF Service chaining represents a path taken by traffic routed through several NFs to benefit from a network service (NS). For instance, a NS composed of Firewall, IDS and Parental control Nfs.
Limits of the existing slicing implementations
In general, the instantiation of network slices is in the form of VNFs via virtualization containers (VM / CNT). The use of virtualization containers induces a virtualization overhead potentially problematic for some IoT deployment targets (e.g. RPi used as IoT gateway) with very limited resources.
Moreover for some NF, by their size, utility, to be instantiated in the form of VNF can be counterproductive. This type of NF is very similar to the anti pattern SOA known as Nanoservice.
Also, this method of instantiation of NF does not cover heterogeneity of the future 5G networks, i.e. the underlying networks will be both classical and cloud-enabled.
The concept of network slicing, as it is conceived currently, is based on cloud-type infrastructure (to allow the deployment of VNFs) and will be hardly usable to achieve end-to-end slices, i.e. connecting data producers and consumers.