Network Slicing: Friend or Foe?

By Tracy Venters
1 Mar 2018

I was looking for some quick stats to start this article by reminding everybody that the internet traffic is growing at a staggering rate. I came across Cisco’s report The Zettabyte Era: Trends and Analysis, updated last June. However, to understand it you really need to grasp what a zettabyte is, as well as an exabyte and probably a yottabyte.

So, I will make it more interesting. By 2021 global IP traffic will increase nearly threefold, traffic from wireless and mobile devices will account for more than 63 percent of total IP traffic, the number of devices connected to IP networks will be more than three times the global population and there will be more than 20 billion IoT devices connected to the internet. Fortunately, broadband speeds will nearly double by 2021 as the Zettabyte Era report predicts it would take more than 5 million years to watch the amount of video that will cross global IP networks each month in 2021. Better get busy binge watching.

As the number of real-time, bandwidth-greedy applications and connected devices continue to grow, networking technologies have had to keep up. For example, Software Defined Networking (SDN) and Network Functions Virtualization (NFV) are just some of the technologies that have helped to create more adaptive networks.

SDN is an umbrella term encompassing several kinds of network technology aimed at making the network more agile. In the simplest sense, SDN separates the control plane from the data plane in the switching network. Instead of the routing intelligence being built into the hardware, it is defined by the network administrator and can be changed as network loads and application requirements change. 

SDN was a promising step forward, however service providers still had a problem. They had data centers full of proprietary hardware to manage. Enter NFV. NFV allows a service provider to consolidate the plethora of applications and specialized network equipment they use in their network onto industry‐standard servers, automatically benefiting from improvements in server horsepower and capacity.

Although not dependent on each other, SDN and NFV are as complementary as peanut butter and jelly, enabling operators to create new revenue opportunities, roll out new services faster while also reducing OPEX and CAPEX.

There is another technology that you may not have heard of as it was only recently made possible through the repeal of net neutrality last December: Network Slicing.

Building on the principles of SDN and NFV, network slicing allows an operator to slice a single, physical network into multiple virtual networks each with its own policies for data speed, quality, latency, reliability and security. Networks can be offered “as a service” with a slice specifically tailored to the unique requirements of a particular application.  

If you Google “network slicing” you will find it is closely associated with 5G. To be clear, network slicing is not dependent on 5G. 5G is a mobile access technology whereas network slicing takes place in the physical network. However, 5G will play a major role in enabling the 20 billion IoT devices and 5 million years of binge watching per month expected by 2020. Many trials of network slicing are already underway between network operators and equipment vendors.

Network slicing allows for prioritization of network traffic. Especially for IoT, some apps are much more important than others. It is more important that a self-driving car detect a red light than send a message to the dealer to schedule an oil change. Total failure of the refrigeration system in a restaurant is probably more urgent than the wait time for a table.

However, it is important to understand that network slicing affects everything that rides over the IP network. It is not just IoT, this could also affect emergency services, calls between remote locations and headquarters, conference calls, data services and much more. Enterprise customers need to understand what it means to them and ensure their agreements with service providers are structured to support their application requirements.

Some of the terms and conditions to be addressed include:

  • Latency – the time it takes for a packet of data to traverse the network
  • Reliability – availability of service without interruption caused by network outages
  • Coverage – availability of a service within a certain geographical area
  • Number of connected devices supported
  • Data Security – both customer and enterprise data privacy
  • Guaranteed Service Level Agreement (SLA) – guarantees of performance with respect to the above listed characteristics and more

Enterprises need to ensure that all of their traffic types are not throttled in favor of higher paying traffic traversing the operator network. The guarantees in the T&Cs may be able to protect against rising carrier pricing.

Don’t wait until 2020. Now is the time to start reviewing your contracts as your provider may already be deploying, or planning to deploy, network slicing on the current network. Make adjustments now where needed to ensure your slice of the pie is everything you thought it would be.


There are currently no comments on this article.

You must be a registered user to make comments