Internet of Things (Iot) plays a crucial role in the development of the telco edge, according to a latest report by technology research firm International Data Corporation (IDC).
Entitled “The Topology of the Edge: Bridging Enterprise Edge to the Telco Cloud”, one of its key takeaways states that communications service providers (CSPs) “will need to build their own cloud-native server solutions and software stacks that integrate into the 5G infrastructure and support a variety of real-time and non-real-time IoT, OT edge, B2C and B2B2X applications”.
IDC views the edge as comprising four broad categories: the enterprise edge, the telco edge, the operational technology (OT) edge, and the IoT edge. The telco edge, by adopting cloud-native technologies, will be able to carry workloads from the enterprise edge, OT edge and IoT edge by spreading and pricing the network infrastructure cost similar to the way that virtualised Infrastructure-as-a-Service (IaaS) is offered by cloud service providers.
“The scaling IoT use cases and adoption of wireless SD-WAN devices, and SDN-enabled transport are key drivers to telco edge infrastructure, going forward. In large telco edge implementations, there will be many edge applications from different stakeholders, which means the telco edge will need to be managed and operated as a multi-tenant edge-cloud system,” says Bill Rojas, Adjunct Research Director at IDC Asia/Pacific.
The report investigates how the evolving telco edge cloud can be bridged to the enterprise edge as CSPs construct their 5G network infrastructure. The telco edge will develop in parallel with 5G Standalone (SA) deployments starting in 2021 that will utilise 5G connectivity types such as ultra-reliable low-latency (URLLC), massive Machine Type Communications (mMTC), and enhanced mobile broadband (eMBB) services.
Telco edge extends life of network infrastructure
The telco edge is located typically near mobile cell sites and/or regional/local data centres, the latter of which might be much smaller than a typical central data centre. In this way, the telco edge is a heterogenous network of far edge and regional data centres. Open source, cloud-native technologies, and industry standards are key to making the telco edge possible.
Also known as MEC, the motivation for the telco edge is to bring distributed cloud computing technologies, digital platforms, and business models together in order to construct a multi-tenant distributed edge cloud ecosystem, according to the IDC report.
It added that while the ICT industry had focused on centralisation via cloud computing to share infrastructure and reduce costs to the enterprise, the MEC is all about distributed computing at the telco edge in order to enable low-latency and high-bandwidth use cases that otherwise would not be feasible with centralised cloud architecture.
“Example of use cases ideally suited for MEC are autonomous transportation, V2X, AR/VR, high-performance gaming, and real-time sensory and image processing,” the report stated.
By distributing the compute and storage resources into the telco edge, the amount of IP traffic flowing back into cloud data centres can be reduced significantly which extends the useful life of the service provider network infrastructure.
A number of CSPs in Asia/Pacific are actively pursuing the telco cloud/telco edge, including Telstra, VHA, Bharti Airtel, Reliance Jio, Vodafone IDEA, Rakuten, SK Telecom, KT, China Mobile, China Unicom, and China Telecom. China alone has over 100 MEC pilot projects underway.
As 5G SA networks go live and network slicing becomes a reality, the telco edge will become an important new source of revenue for CSPs, web-scale companies, such as Microsoft, Google, AWS, Tencent, Baidu, Alibaba, and their ecosystem partners.
In the past few months, a number of strategic collaborative partnerships have been announced in the U.S., Europe, and Asia/Pacific. The business model for MEC is still work in progress, and several different models are being explored including Pay-as-you-go, IaaS, and wholesale models with various performance metrics – peak speeds, Quality of Service, data caps, and etc