Introducing Network Slicing

A short guide to the concepts and use of network slices

Introduction

Concept

The term “Network Slicing” has been defined by numerous standards bodies including 3GPP, IETF and GSMA.

Many standardisation bodies focus on the 5G aspect of slicing.

Zeetta’s definition extends beyond the 5G domain but all views of network slicing have a common conceptual base:

  • the slice is an isolated, virtual network targeting a particular usage of the network

  • the slice is a partition of the network topology and resources

  • the slice facilitates the use of shared network infrastructure for heterogeneous use cases

  • the slice has defined connectivity and quality of service

 

3GPP states

Network slicing is a paradigm where logical networks/partitions are created, with appropriate isolation, resources and optimized topology to serve a purpose or service category (e.g. use case/traffic category, or for MNO internal reasons) or customers (a logical system created “on demand”).

IETF states

IETF Network Slices that provide connectivity coupled with a set of specific commitments of network resources between a number of endpoints

IETF Network Slices are created and managed within the scope of one or more network technologies (e.g., IP, MPLS, optical). They are intended to enable a diverse set of applications with different requirements to coexist over a shared underlay network.” (here)

GSMA states

Network slicing is the embodiment of the concept of running multiple logical networks as virtually independent business operations on a common physical infrastructure in an efficient and economical way.” (here)

End-to-end Enterprise Network Slicing for applications

Network slicing is a procedure to create multiple logical end-to-end networks across multiple technologies and administrative domains.

A network slice is a virtual network, an isolated subgraph of a network topology which consumes a subset of the network resources achieving connectivity, security and a specified quality of service. Each slice is provisioned based on the specific needs of the application that will use the slice, and specified by the network slice definition.

In our system, the network slice definition is provided as a declarative intent giving the access (endpoints) and quality of service requirements for the slice based on the connectivity requirements of the application.

Zeetta's Platform

Zeetta’s end-to-end network slicing platform for enterprises provides the following capabilities:

  • Topology and Resource onboarding

  • Slice Intent Processing

  • Automated Slice Design

  • Automated Slice Provisioning

  • Slice Monitoring

  • Slice Adaption

The solution is based on vendor-neutral network abstraction and pluggable ecosystem integration.

The end-to-end network slice is segmented into network slice subnets. Each network domain has its own network slice subnet designed to support configuration, monitoring and assurance of the network slice across multiple administrative and technology domains.

Use Case Examples

The diagram below shows three network slices for three industrial use cases – “Asset Tracking”, “Push-To-Talk” and “Remote Training”.

The three applications have different connectivity and quality of service requirements but share the same network infrastructure.

Asset Tracking

This application tracks valuable and sensitive assets moving within and between industrial sites.

  • Distributed App
  • Wireless sensors
  • Wide coverage
  • Highly secure

Push-To-Talk

This application provides WiFi and 5G Push-To-Talk capability

  • Distributed App
  • Wireless handsets (Wifi and 5G)
  • Highly available

Remote Training

This application provides remote AR/VR industrial training on robotic equipment.

  • Distributed App
  • Wireless equipment
  • Low latency
  • High bandwidth

Value

The potential benefits of network slicing are significant and cover multiple requirements.

Efficiency

  • Network slicing supports the sharing of network infrastructure in a more efficient way because slices are tailored to applications and this reduces over-provisioning and the amount of network infrastructure required

  • Network slicing encourages automation and this automation when associated with machine learning reduces the operational cost of managing network infrastructure

  • Network slices can automatically scale elastically which prevents costly over-provisioning of connectivity services

Reliability, Availability and Performance

  • Network slicing implements the quality of service required for applications as part of the slice design process. To assure network slices, each network slice can have dedicated, segmented monitoring and slice segments can adapt to network conditions

Security

  • Network slicing provides network isolation in terms of traffic and the network resources provisioned. This feature of network slices supports good network security patterns. Isolation supports slice-specific policies and security rules than otherwise would be much more difficult to manage

Manageability

  • Network slicing achieves an end-to-end multi-domain network slice which helps users understand and adapt their network for their business applications. End-to-End multi-domain slices can be viewed through a “single pane of glass“

  • Network slices are extensible and scalable and can be automatically adapted based on the monitoring data or input from different types of users reducing operating challenges