E-Handbook: What you need to know about edge computing architecture Article 3 of 4

ltstudiooo - Fotolia

What is multi-access edge computing, and how has it evolved?

Multi-access edge computing provides the processing capacity needed to support the increase of 'things' at the network edge. But for all its promise, MEC has challenges to face.

Multi-access edge computing is based on the principle that processing capacity at the edge of the network will provide significant application benefits in terms of responsiveness, reliability and security. Despite the increasing number of vendor options, multi-access edge computing is in its early stages, with many potential buyers in the investigation or pilot phases of deployment.

Multi-access edge computing (MEC) is a network architecture that supports compute and storage capacity at the network edge, rather than in a central data center or cloud location. MEC enables rapid and flexible deployment of new applications, and it offers significantly lower latency -- and better performance -- for local applications and data, compared with centralized data center resources.

Applications for multi-access edge computing

Prototypical MEC applications require ultrafast response times and high availability, and they derive security benefits from localized data flows. MEC provides the intelligence for taking real-time actions and the ability to perform complex data analytics. Applications suited to MEC capabilities include virtual reality, self-driving cars and business-critical IoT applications, all of which require real-time response.

Any application that generates a large amount of data can benefit from MEC, as edge computing can make immediate decisions and only transmit aggregate data to central cloud infrastructure, thus significantly reducing network bandwidth requirements.

edge computing
Edge computing supports applications and devices that require real-time response.

The emerging standard for the next generation of cellular wireless networks, 5G, has already incorporated elements of MEC capabilities in its radio access network (RAN) architecture. By using local compute at the RAN, 5G can power intelligent traffic routing and prioritization. Multi-access edge computing enables coordination of 5G's massive bandwidth capacity and reduces latency for real-time applications, like video.

Status of multi-access edge computing

Service providers will be early adopters of multi-access edge computing for specific high-density locations, such as sport stadiums, and initial 5G buildouts. Organizations with business-critical IoT applications with high data and low latency requirements -- like IoT gateways in manufacturing, for example -- are also good candidates for MEC deployments.

The MEC acronym previously stood for mobile edge computing. In 2017, the European Telecommunications Standards Institute adapted the name of its MEC industry specification group to multi-access edge computing, in a move to include the range of applications and devices at a network edge, rather than referring solely to mobile.

The lack of clear standards and a variety of architectural options are complicating MEC adoption. A number of standards organizations are competing around multi-access edge computing, each focusing on specific aspects of MEC deployment. Some of these organizations include European Telecommunications Standards Institute, OpenFog Consortium and OpenRAN.

No hardware has emerged as the leading platform for MEC, but options include white box Intel servers, hyper-converged platforms, Central Office Re-architected as a Datacenter and ARM. Another challenge is multi-access edge computing lacks a robust software ecosystem due to the diversity of edge application requirements, like the multitude of IoT applications.

Dig Deeper on Network infrastructure

Unified Communications
Mobile Computing
Data Center