The future of digital technology relies on data-intensive and ultra-low latency applications to be developed. To support this, Vodafone Business are betting big on multi-access edge computing (MEC) being central to the development of these new use cases.
“Up until this point, application developers have viewed networks as a fixed variable and would design their use case around the minimum bit rate or streaming feed the network could provide,” says Jennifer Didoni, head of cloud, edge and mobile private networks at the multi-national carrier. “With MEC helping networks to become more of a cloud service that can scale up and down on-demand for specific devices and applications, a world of new possibilities opens up.”
Developments in this space go hand-in-hand with rapid advancements in automation and AI, applications that will require a much more dynamic network. As a result, Vodafone are seeing a host of exciting new use cases taking advantage of MEC capabilities.
What is MEC?
“MEC is a capability that brings cloud computing closer to the end point of a network. The traditional method for customers consuming cloud services is through cloud regions across different countries or through hosted data centres,” Didoni says.
This can help reduce the journey time from the end device to the service provider from 150-250ms, to below 25ms, allowing content and computing to move far closer to the end user.
Edge computing itself is nothing new. Industries such as manufacturing and logistics have been harnessing the power of edge computing for a while.
However, the next step is dedicated MEC, which Didoni defines as “cloud computing resources on a company’s premises, completely controlled by the customer with the data getting to the edge computing device over a private network.” Dedicated MEC has now reached relative maturity in sectors such as agriculture and health, alongside industries that were first to adopt edge computing.
Implementing dedicated MEC can lead to response times of as little as 10ms, which is faster than the human brain can process an image (13ms was the fastest time ever recorded by researchers at MIT in 2014). This ultra-low latency can be used for tasks such as live-remote operated surgery if a hospital was to use this technology. But for Vodafone, the real growth area is distributed MEC.
MEC on public networks
Vodafone are building cloud computing directly into its public 5G networks at strategic points of concentration. “We’re seeing more pilots and proof of concept (PoC’s) than huge scale implementations at the moment, but we’re certainly encouraged by the growth,” Didoni explains. Distributed MEC is the key to unlocking the full potential of the most exciting use cases 5G can offer, from smart cities to autonomous vehicles.
“One of the exciting projects we’ve been working on is with a company called Imperium Drive,” Didoni adds. Imperium Drive remotely drive rental cars to customers. To do this safely, they need to ensure their video feeds have as low latency and are as reliable as possible.
Since 2021, Vodafone have launched distributed MEC services in four countries. In the UK and Germany, they have partnered with Amazon Web Services (AWS) Wavelength as its cloud provider, whereas in Italy they have built its own proprietary software. Meanwhile in Spain, Vodafone launched a pilot zone with AWS for its customers to run pilots and PoC’s.
“We’re still at the early stages of distributed MEC adoption,” Didoni admits “To be able to offer the flexibility that it needs to really complement application designers’ needs, we want to understand exactly how end users want to engage with us.”
They are slightly different propositions as well. The MEC’s pioneered by the Italy team is delivering a business outcome, offering an end-to-end solution that is running low latency applications. Whereas the AWS Wavelength is offering cloud computing infrastructure as a service from a different location within the network.
Didoni considers Vodafone to be in a fortunate position operating in a number of different markets. “This allows us to experiment with how this technology will be adopted so we can ask: Does using a third-party hyperscaler work or is it more convenient if we can offer an end-to-end solution with the cloud computing run by Vodafone as well?”
Right now, it’s still too early to tell, but while A/B testing makes sense for Vodafone in this instance, a lack of uniformity across geographies could be one of the obstacles holding back distributed MEC from further adoption.
Didoni acknowledges that “application developers want to see a consistency in edge services in multiple countries with the ability to provide a seamless service with a guaranteed quality level. They want its solution to work everywhere, not just in specific countries.”
Companies could be waiting for this level standardisation to come to fruition before really rolling out MEC services at scale.
Where is MEC being used today?
“It’s important to remember that MEC is not a replacement for public cloud,” Didoni stresses, before moving onto some specific examples. Vodafone Business is keen to work with its customers to make sure they fully understand which workloads are sensitive to latency, or what data is so sensitive that keeping it on-premises makes an enterprise feel more secure. Core IT and cloud estate may stay as it is in public cloud, so they are very much co-existing rather than competing.
“One of the challenges that we try and help customers with is how to implement MEC where it makes sense but avoid disrupting their other workflows where they rely on the public cloud,” Didoni explains.
Why is MEC such a game changing solution for innovation? Didoni believes “one of the most compelling business cases and probably the one that receives the most attention is the reduction of latency”. One example she points to is a customer that is using a virtualisation application identifying anomalies in its industrial processes. The customer is using a dedicated MEC solution to react in real-time to any issues that arise.
But there are certainly other benefits beyond this, that perhaps don’t get the credit they deserve. As Didoni alludes to when discussing where MEC fits in as an alternative to public cloud, data sovereignty is one such benefit.
“More and more we are seeing organisations far more sensitive to where their data is stored. Using MEC enables data to be processed and stored at the edge of the network, within the country or region where the data was generated, rather than being sent to a centralised location outside of the jurisdiction,” Didoni explains. This can help ensure that data privacy regulations and sovereignty laws are upheld, providing greater control and security over sensitive data.
For applications where the intended use case is collaboration, ensuring a more predictable service can help make the user experience far more natural. Vodafone teamed up with US carrier Verizon and Canadian carrier Rogers, to show that there are green shoots of interoperability in edge compute.
The project they worked on together allowed musicians to play together remotely without delay. The operators “used network APIs into a 5G phone that queries the network, identifies where the nearest low latency edge compute is and interconnects the different operators edges so the app can provide the right buffer for the end users in real time with one another.”
Consistency of service is vital for such a use case and would not be possible with out MEC.
How to drive further adoption
“A lot of MEC use cases are new applications that require a business case and a new approach,” Didoni says. “It might not be something that enterprises are doing currently and replacing, it is more likely to be a new solution or a new dataset they want to adopt to innovate their current processes.”
With this in mind, Vodafone are keen to work with customers to understand the business outcomes they want to achieve and work with them to build infrastructure that can support them.
To be able to support this when Vodafone first launched its distributed MEC site in London they simultaneously launched the Edge Innovation program, where they have invited customers, partners, independent software vendors and universities to sign up and access tools and information on how to construct edge applications.
Vodafone have also run over 40 pilots as part of the program, “providing hands on support from AWS and Vodafone to make sure people have the right SIM, the right device and making sure workloads are moved in the right way.”
This year at Mobile World Congress, Vodafone partnered with the 5G Future Forum, a collaboration among operators to develop 5G interoperability specifications and accelerate the delivery of 5G and MEC solutions around the world on the MEC acceleration program. Like the edge Innovation program, this collaboration also invites ecosystem partners to a pilot network as a service API’s the operators are providing together and to run demos.
“It’s a great time to be a part of the MEC world,” Didoni summarises. “We are approaching a perfect storm whereby networks are evolving to suit the needs of all these exciting use cases just as AI and automation are really starting to take off at scale.”
