For communications service providers (CSPs), the network edge is high stakes. There’s opportunity to create new services, better customer experiences, and more revenue, but there’s also plenty of challenges. Recently, in a repurposed industrial space directly adjacent to the on-off ramp for the 101 Freeway over the San Francisco–Oakland Bay Bridge, five execs – from Red Hat, Hewlett Packard Enterprise (HPE), Intel, Orange and Deutsche Telekom – joined on a panel hosted by TelecomTV to discuss those opportunities and challenges, and give some insights on the future of mobile edge computing (MEC).
To be fair, the edge may mean different things to different people, but for many on the panel, (which was moderated by TelecomTV editor-in-chief Martyn Warwick) the edge is any part of the network that’s not the data center, outside of the core, and closer to the user.
By the way, Wikipedia defines MEC as a network architecture concept that enables cloud computing capabilities and an IT service environment at the edge of the cellular network. Earlier this year, ETSI broadened its focus from MEC to multi-access edge computing (MEAC) in order to broaden the scope to include WiFi and other access technologies. TelecomTV covered the change in this article, writing that a ‘heterogeneous’ edge computing capability is necessary in part because WiFi is now often used interchangeably with cellular, and because "new uses like IoT would need edge computing as a filter to stop the core network being overwhelmed by data.
The panel spent a fair amount of time discussing use cases, like IoT, that will warrant the CSPs investing more in edge computing. Other use cases mentioned include virtual reality (VR) and augmented reality (AR), gaming, emerging new apps and services within the auto industry, and more.
"In addition to those use cases, if interfaces are kept open there could be this tremendous remixing of technology so you can get totally new use cases," said Sanjay Aiyagari, panelist and senior solutions architect for telecommunications & NFV with Red Hat. "Think about the health care industry. There’s a need to monitor different kinds of devices, but that could require integration with so many different systems. Then it’s not just one use case, and developers could mash-up location-based services with sensor data and with other things. Put it all together, and it could be really interesting."
The panelists also discussed the edge computing’s value chain, how it will be monetized and what the competitive landscape looks like for the CSPs (several referred to the edge as the new battleground between the telcos and the webscale companies), how NFV is evolving and the role it is playing in the facilitation of new IoT and 5G services, and where open source fits into edge computing.
In addition to Red Hat’s Aiyagari, the panel featured Alex Reznik, ETSI MEC chair and enterprise account architect with HPE; Keate Despain, director, Business Acceleration Network Platforms Group, Intel; Christos Kolias, principal research scientist, Orange; and Axel Clauberg, VP, Aggregation, Transport, IP, (CTO-ATI), Architecture, Deutsche Telekom.
There’s a lot more on edge computing and the opportunities and challenges for CSPs, so take a few minutes to watch the video of the panel.