Saturday, November 25, 2017

How telcos and data centre operators will embrace the next generation of edge computing

Much has been said about edge computing, however, beyond the hype there are some critical issues that need addressing. João Marques Lima talks to Red Hat’s Ian Hood on how telcos and data centre operators will embrace the next generation of proximity computing.

The most recent Gartner’s “Hype Cycle for Emerging Technologies” shows that edge computing is on the verge of becoming an “innovation trigger” with mainstream adoption expected by the early 2020s.

Within this new edge world, there are many considerations around interoperability of edge sites and central data centres, workflows, reliability, availability, and security to name a few.

In addition, the topic of mobile edge computing is also emerging and its impact on telecommunications and data centre operators is only now starting to be addressed in depth. With that in mind, Data Economy spoke to Ian Hood (IH), chief architect, global service providers at Red Hat, to learn more.


What does MEC have to offer to telcos? 

Ian Hood, chief architect, global service providers at Red Hat

IH: Mobile Edge Computing offers many things to telcos. Because service providers host edge locations around the world, they interconnect with connected devices. Therefore, they are very well positioned to support the IoT market by bringing intelligence to the network’s edge in the form of MEC.

As networks are becoming increasingly virtualised, MEC offers telcos an architecture on which to develop new services around IoT: there are opportunities to offer innovative new services and experiences by delivering them closer to their customers.

In addition, use of MEC architectures enables operators to consume technologies from many vendors and optimise the efficiency and utilisation of their infrastructure resources, while accelerating time to revenues.


What sort of next-generation services will MEC create? 

IH: MEC enables many types of use cases for the operators: IoT services across all industry verticals, virtualisation of their radio-access network (RAN) as part of their LTE to 5G architecture evolution, Enterprise/ Retail Edge, and distributed content/ video delivery services.

Some operators are already looking at how they can offer Artificial Reality/ Virtual Reality as well as the use of Artificial Intelligence in delivery of innovative use cases.


What security hurdles exist today around MEC? 

IH: MEC can leverage many emerging technologies such as Software-Defined Networking (SDN), virtualisation, containers and microservices, to build a cloud-native distributed network that can enable the deployment of any application over any infrastructure.

Many of the same security hurdles exist for large distributed networks as we begin the journey to 5G/ MEC architectures, these include encryption, DDOS, privacy, lawful intercept, identity and role-based access.

I would recommend that an organisation running its applications over a network implements security precautions for those applications, regardless of the infrastructure underneath.

Security measures should include role-based access control for people using the applications, as well as encrypting the application before it even gets on the wire, using SSL and other methods to provide more secure data transfer.


What is the role of data centre operators in the MEC revolution?

This article originally appeared in the last issue of the Data Economy Magazine. Click here to read more.

IH: I anticipate that the MEC revolution will expand the number of places where data centre technologies are deployed, such that they are right alongside the networking technologies and in some cases right back on the enterprise/customer premise.

As a result, the data centre operators and the network operators should take advantage of DevSecOps/Agile approaches to develop their infrastructure/environments as code, to enable them to operationalise this hybrid cloud native distributed architecture.

These approaches can help them offer “always on” services/applications wherever their customers need them, accelerate the pace of changes in their offerings, and to minimize the complexity of these applications/ services deployments with innovative automation, policy, and analytics tools.


How fast is demand growing for MEC and what do you predict for the near future?

IH: According to Market Research Future, ‘the global The Edge Computing Market is expected to grow at ~$33.75bn by 2023, at ~35.0% of CAGR between 2017 and 2023,‘ indicating that edge computing is likely to see healthy growth over the next five years.

I believe this is indicative of the growth we will see specifically for MEC and is pretty much aligned with the interest to deploy 5G architectures to expand operator revenues while optimizing their operational cost structures.

While operators continue to expand use of SDN and Network Functions Virtualization (NFV) to accelerate the delivery of innovative services, they are already starting trials of many of the MEC use cases now to validate the maturity of the emerging technology advances and the cost benefits they expect to gain.


What sort of other technology trends is MEC already generating?

IH: Another key trend that MEC is generating and expanding is the need to modernize both IT and networking applications for web-scale, cloudnative deployments.

The trend towards microservices architectures that modularise legacy software applications into much smaller more granular elements is key to being able to cost effectively deploy MEC applications at scale.