Why “network slicing” is the key to 5G network adoption
Bandwidth. Digital gold. Internet lifeblood. Jugular veins of international connectivity. Selfie distribution system.
You get the idea – bandwidth is something everybody wants, especially mobile bandwidth! More bandwidth is the most coveted benefit people commonly refer to when discussing tomorrow’s 5G mobile networks, and why wouldn’t they? The 5G bandwidth promise is a whopping 10Gbps to each end user, although is best case and likely use case-specific. Users addicted to their smartphones will insist that they “need” 10Gbps of bandwidth, but what would someone actually do with this amount of bandwidth? Like, how many high resolution cat videos can you watch at once? How big is your smartphone screen?
Well, this is the foundation for network slicing – rightsizing mobile connectivity for use cases.
So, what is network slicing, anyway?
Technically speaking, slicing allows Mobile Network Operators (MNOs) to manage and operate multiple virtual networks over a common physical network infrastructure. It enables the virtual partitioning of the Radio Access Network (RAN), core network components of the Evolved Packet Core (EPC), and switching and aggregation network all the way to the data centers, where accessed content and applications are hosted.
An end-to-end service orchestration engine “slices” virtual (logical) networks out of physical network resources to enable mobile Network-as-a-Services (NaaS) specifically tailored for each use case. Slices will be characterized by specific use cases that dictate the required network performance related to such metrics as capacity, latency, security, duration, reliability, and geographic coverage, to name a few.
More use cases, more revenue streams
5G promises to support a broad and varied number of use cases, many of which are expected to justify and fund the rather expensive buildout of this next generation of mobile network. This has resulted in a variety of aggressive 5G aspirational performance goals, as mentioned in an earlier blog, and listed below.
- Up to 10Gbps connection rates to mobile devices in the field
- Up to 1000 times increase in bandwidth, per unit area
- Up to 100 times more connected devices
- A perceived network availability of 99.999%
- A perceived 100% network coverage
- Maximum of 1ms end-to-end round trip delay (latency)
- Up to 90% reduction in network energy utilization
The aspirational goal that gets the most attention in the media is the first one – 10Gbps to the mobile user – likely because it is something that we can all appreciate firsthand. However, the other performance goals are equally important to many of the new use cases expected to drive 5G adoption. The following use cases are often touted as the reason why 5G mobile networks are required, and vice versa.
- Fixed broadband residential internet access
- Autonomous self-driving cars
- Improved user mobility (ex. high-speed trains)
- Broadcast communications services (ex. mobile television)
- Extreme real-time communications (ex. tactile Internet, augmented reality, virtual reality)
- Critical lifeline services (ex. first responder services)
- Ultra-reliable communications (ex. remote eHealth monitoring)
- Massive Internet of Things (ex. Internet of Cows, yes this is indeed a real thing)
5G use case marketing challenges
Migrating services that currently use their own standalone wireless networks, coupled with creating new services that will make use of 5G mobile network performance, will help build the business case for the rather significant investments that MNOs will have to expend to deploy the required new 5G technology. Getting users to adopt new services that can’t be served by existing 3G/4G mobile networks may prove far easier than convincing users of dedicated wireless networks to switch to 5G networks. This is because there’s a good reason why they have their own dedicated network, which includes control, customization, availability, reliability, security, familiarity, and other attributes optimized to their specific use cases that would have a hard time, if even possible, being served by existing best-effort mobile network services.
For example, let’s look at first responders who we rely on to save our lives when disasters occur, such as fires, earthquakes, tornados, hurricanes, and increasingly (and sadly), those that are intentional. These brave people often make use of communication services that are served by a dedicated wireless network, allowing for the coordination of activities to best serve those in need. Having these users move off their tried and trusted wireless network that’s been customized for decades to their specific needs will surely be a formidable task indeed for MNOs. Cost savings alone simply won’t convince them to abandon their dedicated wireless network and migrate to a best effort mobile network shared by everyone else – unless network slicing is implemented, which will deliver the same performance over a shared mobile network.
5G network slicing will allow MNOs to offer virtual networks over the same physical mobile network used by everyone else that will mimic the same or better expected network performance as the replaced dedicated network in terms of capacity, speed, latency, and most importantly, availability, in the case of the first responder example above.
For instance, as parts of the mobile network goes down due to a hurricane knocking down radio towers, first responder communications services would be rerouted with a very high priority over everyone else, albeit likely at a price premium. This would ensure that the highest network access priority be given to the firemen communicating with each other at a disaster site and very low priority to the people taking selfies at the same disaster site. Slicing will allow for this prioritization.
Virtualization is key
To rationalize significant investments and ensure long-term financial viability, 5G networks must support a wide range of revenue-generating use cases, which will require different network service performance. A use case approach will help to ensure that the new 5G mobile network remains highly profitable for the MNO, else why bother to upgrade the network in the first place? To achieve these goals, 5G networks will have to be adaptable, dynamic, and programmable from end-to-end using virtualized constructs. This is facilitated by implementing network slices for each supported use case with tailored performance invoked autonomously and programmatically. A fully flexible end-to-end 5G mobile network will allow MNOs to offer a far broader range of communication services built on existing, new, and yet to be created use cases that will ensure innovative and ongoing revenue streams beyond simple connectivity and capacity.
To address the plethora of possible existing and new use cases in a timely and cost-effective manner, 5G mobile networks will likely be based on logical (virtualized) building blocks rather than traditional physical building blocks. This will reduce the time-to-market for new, innovative service offerings allowing for first-mover competitive advantages. Anything that increases differentiation of mobile network services will surely be looked at closely by MNOs, as the commoditization of mobile bandwidth is already upon us.
SDN and NFV building blocks
The many promised benefits enabled by SDN ultimately involves the complete abstraction of the physical network infrastructure. This allows network behavior to be adapted to the required guaranteed service performance required of each use case supported. Network slices can be implemented that are tailor-made for each of a broad range of supported use cases over the very same pool of physical network assets. This allows MNOs to deploy a single physical network capable of supporting a broader service portfolio to diversify, expand, and increase ongoing revenue streams in a highly cost-effective manner. Gone will be the days of a single mobile network infrastructure that for the most part offers best effort services to all users, which sorely limits the amount and type of use cases enabled and supported.
NFV allows network functions to be deployed and implemented strictly through software rather than buying, shipping, installing, and turning up physical appliances. One of the most promising benefits of NFV is the ability to execute a variety of Virtual Network Functions (VNFs) independent of physical location, meaning VNFs can be deployed and executed in different parts of the greater network for each network slice, wherever it makes most sense related to performance, cost, reliability, availability, and security.
Security concerns and challenges
5G network slicing is predicated upon the sharing of the same underlying physical network infrastructure by logically separating different use cases. This means end users must be very comfortable that their network traffic is fully and securely isolated from others, especially for some of the newer use cases, such as the first responder use case discussed above. The security challenge will involve proper technology selection and implementation, as well as successfully addressing the concerns of the more paranoid. Convincing all users of all use cases to share the same network will undoubtedly be a challenge for MNOs to overcome, but it’s doable if the 5G network is properly designed, sliced, and secured for different use cases. However, implementing network security technology is one thing, but convincing some users that their inflight data is secure is quite another. It’s possible, albeit with big investments in time (and money).
MNOs want to deploy, manage, and offer services at the lowest possible cost without compromising the quality of service or experience of their customers. This means that 5G networks will have to be managed in an automated and orchestrated manner from a service provisioning perspective. Orchestration of network service connectivity, coupled with daisy-chained network functions, will require multi-domain and multi-vendor management support. Each 5G slice will entail tailored use case network performance delivered via individual physical and logical nodes from a variety of different vendors. This is exactly where the concept of openness will help MNOs more quickly and easily implement 5G network slices and associated network services. Open, standardized software interfaces coupled with Multi-Domain Service Orchestration (MDSO) will be key to use case-centric 5G network technology and business models.
Slicing up the 5G network pie
It should be noted that the 5G concept of virtually “slicing” shared physical network resources has been around for decades in the form of Virtual Private Networks (VPNs) over packet-based LANs and WANs.
What’s old is new again, right? Well, not exactly.
5G network slicing will surely be far more involved and challenging than traditional VPNs because it will involve the combination of many different types of existing and emerging networks technologies related to wireless, wireline, RAN, EPC, SDN, NFV, optical, small cells… that’ll be associated with far more demanding use cases, many of which haven’t even been dreamt up yet. Regardless of the formidable technological challenges of network slicing, it’s seen by many as the best way to justify 5G investments, as it enables a wealth of existing and new use cases that justify 5G investments today and into the future.
References: “Understanding 5G - Perspectives on Future Technological Advancements in Mobile”, December 2014 (GSMA Intelligence)