Network Slicing and Private LTE/5G
Network slicing is widely recognized as a key feature of 5G networks that will be used to address enterprise and industrial wireless requirements. At the same time many enterprises have been deploying business-critical Private LTE/5G networks and experimenting with Private 5G.
One of the hottest topics in 5G relates to its digital enterprise applications, from the industrial factory floor to hospitals and many more. In fact ABI Research recently stated that Private 5G will outrun public 5G for spend and spectrum.
Most of the 5G material we find on the Internet tells us that the novelty of 5G is not simply its new radio (NR) which provides the obvious benefits of higher speeds and low latency, but the 5G Core that plays a central role. For those who are less familiar to mobile network architecture, the 5G network can be broken up into two main components: radio and core. It is the 5G Core that controls the network and therefore becomes a key enabler for vertical market applications. The 5G Core features (and where they reside) are therefore very important.
There is a lot of marketing describing what these 5G Core features are. Most would mention “Network Slicing”. But what is network slicing? For those who are new to the term, network slicing is a technique introduced in 5G where the virtualized wireless network infrastructure can be logically subdivided into network slices each of which satisfies a specific use case. For example, a network may be sliced so as to set aside a “radio slice” a “core slice” and an “IP/transport network slice” (connecting radio, core and the enterprise Intranet), each with a given amount of capacity and characteristics (speed, latency etc.), that can be combined to produce a dedicated service for e.g. connected cars.
GSMA’s paper on network slicing puts it clearly: “Not a network slice for each vertical Customer”. The assumption is that most enterprises will be happy to share business-critical network infrastructure and functions with their competitors. The logic is that “operational isolation” is sufficient for the majority of customers because “vertical customers could have independent monitoring, control, configuration, or even full operation capability of the network slice”. Instead, “network level isolation, [which] means that vertical customers do not share network function or resources with the other customers” is not considered to be a fundamental requirement and is not addressed by network slicing. Therefore network slicing assumes that only a small number of slices will be required in a national network and that they will be shared by enterprises having similar needs.
While such a form of “shared” network slicing (without network isolation) may seem feasible for some specific use cases that require national network coverage (say connected cars), the emergence of Private LTE and Private 5G has shown that many enterprises do not want to share any hardware or software infrastructure with other enterprises, let alone their competitors. They may be forced to do so in some cases for lack of alternatives or due to the economics (which can be solved with Athonet’s low cost-of-entry hybrid cloud solution) but not when it comes to their business-critical private networks. These enterprise scenarios include manufacturing, airports and ports, utilities, financial institutions, logistics, hospitals and more. It is not just public safety. Sharing network functions makes it impossible to have full control or to customize them for their specific needs. Also, even if the traffic from different enterprises sharing the network function is separated, any problem caused by one customer could affect all the others. In fact, such customers had been deploying WiFi in the past, even if they may have preferred LTE, because they could “own” their private network and have full control. “Our private data traffic needs to stay local and never leave the enterprise network” is the mantra. In the new era of Edge Computing and Virtual Private Clouds, we are likely to see some customers shift to new operational models. However, the networking will remain dedicated to each business-critical vertical customer, not shared as in network slicing.
Network slicing should not be confused with Private LTE/5G as they have different applications in practice. Private networks (manufacturing plants, airports and ports, utility production and distribution sites, financial institutions etc.) will require Dedicated Core Networks like Athonet’s Connectivity Platform (ACP). A Dedicated Core Network provides the enterprise with full network isolation, bringing greater control, reliability, deterministic quality, as it is not shared with other customers. Instead network slicing caters very well to applications such as nationwide fleet management, providing a reliable and controlled service wherever the device is located.
In Private LTE and 5G, the radio network can use dedicated base stations and spectrum (isolated RAN) or it may consist in a “radio slice” from an operator base station that is located within the enterprise location or sufficiently close to it. That is most likely where the Private LTE/5G Core will meet network slicing.
Such solutions can be deployed already today both in 4G and 5G using either “open spectrum” such as CBRS GAA in the USA or using Communication Service Provider spectrum including CBRS PALs that have been recently auctioned.
The bottom line is that both network slicing and private networks will be important for next generation mobile networks, but they have different applications. At Athonet we can support our customers with both.