Why is 5G network slicing the throat of IoT applications?

In IoT applications, 5G network slicing can provide quality services and enhance reliability and security.

5G is a next-generation mobile communication network technology that can provide many key advantages for IoT applications, including low latency, high bandwidth, and support for long-term, dense network coverage indoors and outdoors. Many companies seem to have realized the huge application potential of 5G in the future. In mid-2018, in a 5G use case and adoption survey conducted by research firm Gartner, 66% of organizations stated that they plan to deploy 5G next year (2019), of which 59% said they see the Internet of Things as the main use case, with the main purpose of Operational efficiency.

Another important advantage of 5G is network slicing, that is, mobile network operators can divide their physical network into slices that can be allocated to a single company or ecosystem for their exclusive use. All parties who include traffic in a slice can obtain service quality and reliability guarantees, such as ensuring that the delay of an industrial IoT application does not exceed a certain number of milliseconds. The ability to isolate traffic in a slice can also translate into enhanced security.

This is a way to build a network that supports multiple categories of services in an efficient manner on the same network. Service providers don't build their own private networks just because of ultra-low latency, so they need to build their network in some way by slicing to meet these needs.

Network slicing is just one of the few options for providing enterprises with 5G products. There are different ways to establish a 5G architecture for an enterprise: a private network can be built for the enterprise, or a service provider can run as a dedicated network service on the client, or it can be set up as a dedicated part of the public WAN.

For companies looking to leverage 5G to build their IoT architecture, the grim news is that it is unclear how long 5G public network connectivity and coverage and 5G network slicing will be open to enterprises. Gartner research shows that when mobile operators initially build 5G networks, they will focus more on the consumer mobile broadband market than invest in advanced network concepts, such as edge computing and network slicing to meet the needs of certain businesses and industries. This means that companies with rapidly evolving IoT architecture needs may have to wait a few more years to fully own their 5G network.

Although large-scale 5G network slicing may take several years, the good news is that network slicing is not an architectural concept that absolutely requires 5G. It can be used for 4G or 3G, and it can be done without 5G, because it is a universal network concept. However, mobile operators may not yet see a large demand for network slicing in the market, and with the advent of 5G, the expected application development can truly meet this demand.

Another way to think about network slicing is to consider it as a product of a mature virtual private network model (VPN) and network virtualization trends.

VPN coverage is usually limited, and network slicing can extend a dedicated path on the network, from the client to the core of the mobile operator's network, which is not possible with a VPN. In addition, network slicing can isolate services from each other, so that users can allocate resources to services, and these services cannot be used by or affected by services on other network slices. It's like running multiple different networks on one physical network.

Network slicing can be wirelessly connected to the network, transmission network, or mobile core. If mobile operators want to guarantee the quality and latency of specific types of traffic (such as IoT applications), they can be performed in a wide area. These parts. Network slicing starts with client devices, which can be IoT devices, and then become dedicated channels on the mobile air interface. It will continue to be part of the transport network all the way to the core.

Service providers can deploy several different network slices on their core and wireless access networks, dedicated to different types of traffic. Some vendors with a large number of customers in specific vertical industries can create slices for each vertical to isolate mission-critical traffic from other verticals. The largest enterprise customers may get dedicated "hard slices" of the network, and such slices may contain many "soft slices" for the different services and traffic types of the enterprise. For example, one slice is for ultra-low latency traffic, one slice is for high-bandwidth traffic, and the other slice is for services that require extremely high resiliency.

With the development and evolution of IoT applications and other types of services, such as high-resolution video, augmented reality, and virtual reality, the number of network slices that network operators can support in their networks may increase. Creating too thin network slices may reduce the quality of the services they provide, while sharply increasing the number of network slices may pose network management challenges for network operators.

However, the reality is that it may take years for network operators to deal with these issues. Demand for network slicing is currently low, and operators may still need to develop business models and pricing for their slicing. After all, dedicated network resources are not cheap for businesses. Different network operators also need to seek interworking protocols for network slices in order to achieve the purpose of transmitting across multiple networks when needed. In addition, as Gartner points out, as mobile operators move on to the next game-changing architecture evolution, network slicing is not their top priority.

An architecture evolution

Although network slicing is feasible today, 4G and earlier generation networks are considered the "hardest" networks, and the characteristics of 5G-high speed, low latency, and wide connectivity, mean that it can be designed as Support for higher quality and guaranteed services. The evolving enterprise and industrial IoT environment may be a driver for the emergence of applications that require richer services with lower latency, which in turn will motivate more operators to create network slices.

However, with the advent of 5G, it will merge with other evolving technologies and network architecture concepts such as NFV, SDN, cloud computing and edge computing, and changes in base station design, which will give mobile operators more consideration.

5G gives operators the opportunity to rethink how to build their networks in a more efficient way.

Network slicing is a step beyond VPNs, which also makes it sound like a less distant conceptual relationship, such as Network Functions Virtualization (NFV), which allows virtualized network resources, and software-defined networking (SDN), which Separate control plane functions from data plane functions to enable more centralized, flexible, and programmable network management.

NFV and SDN are expected to be more integrated into mobile operator and enterprise network architectures, as they can help these companies expand and manage their networks in a more cost-effective manner. With the development of network slicing technology, it can be implemented independently of any one technology, but it can also be implemented by using the same principles of both technologies. For example, if the slice is extended to the operator's core network for expansion, NFV and SDN can be used for network resource allocation and control, which will also require operators to implement end-to-end service coordination in their networks.

At the same time, network slicing will also appear in the 5G world, and the core and edge locations of mobile networks will become a more flexible concept. With the Internet of Things, more data calculations and analysis can be performed on edge devices, edge clouds, or mobile network towers. In addition, base stations are evolving from a physical unit to separate radios from baseband components. Radios are being distributed closer to users, further redefining the edge.

Today, the edge of the network may be in different places, from the radio to a very centralized core network, everything. If there are 50,000 radios in the network, there may be 50 core sites, so one core site can provide 1,000 radios. As network slicing plays a role in this mobile architecture, operators will need to determine the best place to deploy a gateway to handle different applications.

At present, network slicing is still at the conceptual level. How it quickly becomes a network architecture also depends on mobile network operators. Gartner said that enterprises that expect to use 5G to solve IoT delays may achieve this goal faster if they entrust technology vendors or network operators to build 5G dedicated networks. If companies choose this approach, they must also understand that private networks are expensive and difficult to manage, and may require spectrum licenses. In addition, they may be best suited for campus environments where IoT traffic and applications are not spread across a wide area. For enterprises looking for a wider range of solutions, network slicing may be the best choice, but how long do they need to wait? iJUNCO will actively track and provide the most advanced IoT application technology for the majority of enterprises.