The rising importance of edge computing is both a major opportunity and a major headache for mobile operators.
On the one hand, the distribution of cloud resources – processing, storage, memory and services – to the network edge and even as far out as the device, plays to their natural strengths. MNOs understand how to plan, route and optimize highly distributed networks with large numbers of often moving end points, and the edge model puts greater weight on the connectivity – often wireless – than a heavily centralized cloud platform.
In the central model, they run the risk of the connectivity between the cloud and the devices being commoditized. In the edge cloud, there is the potential to integrate those edge-based resources with wireless networks – especially small cells – and put the MNO in charge of delivering services and new user experiences which take advantage of proximity to the user, from personalized marketing to low latency IoT applications.
However, the headache comes from the immaturity of the platform and its potential fragmentation. An MNO-centric edge cloud would require an operator to plan and deploy large numbers of small cells in an integrated way with other resources like processors, at a time when that is logistically complicated, and when there are no clear standards for linking all the pieces to one another and the cloud.
Yet if the MNO holds back, there is the danger that other organizations will seize the initiative, perhaps harnessing less telco-centric approaches to the edge cloud, such as OpenFog, which distributes cloud resources to every point in the network, from cloud to device, rather than shifting everything to the edge. This approach, and others such as Dell’s Open Edge Computing or Microsoft-backed cloudlets, seem more natural tools for the webscale and enterprise services providers to shape the edge cloud in their particular image.
The news that the IEEE is working with the OpenFog Consortium on standards will sound several warning bells for MNOs. Combining the innovation, but potential anarchy, of an open source initiative, with the processes of a venerable standards body, is what operators say they want – the best of both worlds. But the IEEE is mainly concerned, on the wireless front, with unlicensed spectrum and has closer ties to data centers than telcos, so that could be a real threat to the MNOs’ chance of taking top dog role in the edge cloud value chain.
Already, alternative operators are looking to combine unlicensed or shared spectrum with small cells to create localized, often virtualized subnets – self-contained mini-RANs with their own virtual cores, which can enable optimized services for a certain enterprise, city or vertical industry. Cablecos, industrial IoT players and private network operators could all take this approach, and then add OpenFog or MEC to add further service value – leaving the MNO mainly as a provider or wide area roaming.
And the OpenFog group is in no doubt that it wants its architecture to move well beyond its enterprise roots to be an integral part of 5G as well as the IoT.
Its OpenFog Reference Architecture will form the basis for a new working group formed by the IEEE Standards Association (IEEE-SA). The first meeting of this new group will take place in November and its work should be completed by April 2018.
“This represents a giant step forward for fog computing and for the industry, which will soon have the specifications for use in developing industrial strength fog-based hardware, software and services,” said John Zao, chair of the group, whose full title is the IEEE Standards Working Group on Fog Computing and Networking Architecture Framework. “The objective from the beginning was that the OpenFog Reference Architecture would serve as the high level basis for industry standards, and the IEEE is looking forward to the collaboration in this effort.”
The reference architecture is a technical framework which provides a structural and functional prescription for an open, interoperable, horizontal system architecture for distributing computing, storage, control and networking functions closer to users.
“The mandate for fog computing is growing stronger, driven by the recognition that traditional architectures can’t deliver on the operational challenges for today’s advanced digital applications,” said Helder Antunes, chairman of the OpenFog Consortium and a senior director at Cisco, which initially came up with the concept.
The Consortium itself predicts that the largest market for its architecture, by 2022, will be energy and utilities, followed by transport and healthcare. According to a report it commissioned from 451 Research, hardware will make up the largest portion of fog revenues, at 51.6%, followed by applications and then services, but the mix will gradually shift away from hardware as fog functionality is incorporated into existing devices.
At the opening of the inaugural Fog World Congress this week, Christian Renaud, IoT research director at 451, said: “It’s not only a technology path to ensure the optimal performance of the cloud-to-things continuum, but it’s also the fuel that will drive new business value.”
But there are hopes that MEC and OpenFog may come closer together, rather than supporting two separate and even hostile camps. In September, the ETSI MEC Industry Specification Group (ISG) and the OpenFog Consortium agreed to work together to reduce technical overlap in their work across many domains, to work on APIs to support interoperability, and to share their work in order to drive global standards for a “fog-enabled edge”. However, there was no mention of MEC in the IEEE announcements about standardizing the edge.
Nonetheless, momentum behind the ETSI framework (now renamed Multi-access Edge Computing) has rebounded in recent weeks after a period of backlash, and there is particularly strong support in the influential Chinese market. Many MEC trials have taken place in China, including smart city applications like ZTE’s smart parking system.
At the SDN/NFV Congress earlier this month, China Unicom said it planned to invest in
thousands of mini-data centers at the network edge over the next few years, to enable new 5G services. It will also deploy “hundreds” of regional facilities and “tens” of new core data centers, said Xiangyan Tang, the operator’s chief technologist. He believes this network of data centers will enable 5G – which Unicom will start deploying from 2020 – to support a wider range of services and revenues, including in factory automation, robotics and connected cars.
The Chinese vendors are also highly active in pushing MEC forward. Earlier this year, ZTE proposed a system architecture which it dubbed ‘Cloud-Fog Collaboration’ to “unify the concepts of fog computing and MEC, avoiding the bottlenecks of heavyweight cloud computing.”
It submitted its architecture to the ITU-T CTO meeting in Thailand and showed use cases involving online video, augmented reality/ virtual reality (AR/VR), large-scale IoT and other application scenarios.
In July, fellow Chinese giant Huawei spearheaded a group of 13 organizations which formed a “collaborative circle” to drive the MEC ecosystem. This is heavily driven by Chinese companies, showing the leadership position that the country is taking on the road to 5G architectures, and in devising new service models which it can deploy at home and also export round the world.
All three Chinese operators are in the group, as well as the China Communications Standards Association and China Computer Federation. Also joining were ETSI itself, the GSMA, 3GPP, Industrial Internet Consortium (IIC), Intel, ARM, Trend Micro and iQiyi.
“Currently, there is no industry alliance to promote development of the MEC industry, which is a key challenge for us,” said Long Jiping, VP of Huawei’s cloud core network product line. “All industry players should work together to drive the MEC industry forward and explore edge network capabilities to commercialize MEC and achieve business success.”
Tang Xiongyan, CTO of China Unicom’s Network Technology Research Institute, added: “It is time that MEC applications be developed and used. We need to speed the development of an MEC ecosystem alliance and related standards, while also considering new business models. The goal in all of this is to enable the communications industry to support more industrial internet applications.”