People Consultant, Antonia Basca, and Cloud Transformation Architect, Fragkiskos Sardis, at KPMG offer an informative overview of edge computing and its benefits.

What is edge computing?

Edge computing is a distributed computing paradigm that moves processing and data closer to the consumer. The main benefits of edge computing include decreased latency, higher bandwidth availability, and fine-grained control of Quality of Service (QoS). A few examples of use cases include applications such as industrial control loops, robotics, cloud gaming, augmented and virtual reality (AR and VR), artificial intelligence and virtualised radio access networks in 5G.

How does edge computing work?

Edge computing is implemented in two ways:

  • Standalone servers and other appliances: where one or more devices are used independently to processes workloads without being part of a virtualisation platform. Devices may be shared by multiple users and host several services each or can be purpose-made appliances that perform a specific function
  • Edge clouds and micro clouds: where a cluster of servers is deployed as part of a private or public cloud and centrally managed by one entity that allocates virtualised resources to ‘tenants’ of the platform, with each tenant deploying their own workloads according to their needs

Depending on the use case, the definition of what constitutes edge processing varies and may be a reference to local processing on appliances (such as cameras with embedded AI solutions for object recognition), a data centre local to a site of operations (such as on-premises servers maintained by an organisation), or data centre that provides services for a whole city.

Edge cloud and hybrid cloud deployments

Edge clouds can be part of a larger hybrid cloud deployment, where workloads are divided between the edge and remote data centres or cloud service providers. This is typically done to meet a diverse set of performance criteria and platform feature requirements for specific use case.

For you

Be part of something bigger, join BCS, The Chartered Institute for IT.

For instance, in a manufacturing setting, data that requires immediate processing for automation is handled by the edge cloud before being sent to a remote cloud for further processing and analytics. The advantage of such a deployment is that the edge cloud will only operate on data to issue instructions in real time for controlling industrial equipment, while the remote cloud, which offers a higher capacity of compute and storage resources can be used for long term storage, processing, and analysis of the data.

Edge computing applications

Edge computing has applications in a variety of industries and scientific fields, including healthcare, manufacturing, arts, transport and energy to name a few. Some recent examples of edge computing applications include:

The future of edge clouds

With the market predicted to grow rapidly in the upcoming years, edge cloud is expected to have a promising future. The combination of 5G and 6G with edge computing is the subject of extensive research to enable future use cases:

  • Metaverse and gaming: AR and VR become more popular, more realistic, and vivid experiences are being created. To support sophisticated graphics and broadcast material to numerous users simultaneously across the cloud, the greatest connectivity possible is needed combined with edge computing capabilities to distribute the workloads, localise traffic and provide the ultra-low latency and high bandwidth required for building immersive worlds
  • Human-centric networks: human-centric services further leverage 6G and edge computing as new use cases are developed and the ability to transmit sensory information such as haptics is matured. The ability to see, hear and touch real objects remotely, or virtual objects inside the Metaverse, will further drive the demand for edge computing and enable breakthroughs in arts, healthcare and manufacturing
  • Artificial intelligence: edge cloud is already considered one of the enabling technologies for several AI applications, such as computer vision and robotics. As AI technology matures, it will be possible to launch autonomous drones to collect and transmit information to an AI which can assist in the coordination of robots and humans in activities such as construction, rescue operations, exploration and manufacturing