5G is enabling an era of user-centric network by bringing in the concept of network slicing and composable networking. This has caused a fundamental shift from “network aware application design” to “application-centric network provisioning”. It marks a major step in network architecture. The big question, however, is how the network will compose and organize itself to deliver diverse user requirements that may contend for the same network resources with conflicting attributes. Edge computing is one of the key enablers that helps achieve this goal.
Need for edge computing
Cloud computing has revolutionized the way we perform computational tasks, be it hosting an analytics application, web services or enterprise applications. However, cloud architecture gets challenged when computing servers and high-volume data sources are far away. If the use-case can wait for remote data to be transported within an acceptable time, this architecture still delivers value. When computation needs to be performed in real-time but the data cannot be transported in real-time, then this architecture fails. This is where edge computing comes in. Ideally, equipping the data source with the required computing power is the best option, but it may not be the most economic option. For example, if we want a mobile phone to perform video analytics using machine learning algorithm with a large image repository, it will be a very expensive and power-hungry phone.
Multi-access Edge Computing (MEC) is a natural development in the evolution of mobile base stations and the convergence of IT and telecommunications networking. MEC enables new vertical business segments and services for consumers and enterprise customers. Edge computing architecture brings in computing at the edge, which deals with local data, does the necessary processing to derive insights, and reaches out to the central cloud computing platform with low amounts of data, only if required.
Edge computing is not just about hosting them close to end users and the data source. It is also about the way use-case realization is made possible by fundamental architectural constructs at the edge. MEC provides a new ecosystem and value chain as they have the option to authorize third parties, allowing them to flexibly and rapidly deploy innovative applications and services toward mobile subscribers, enterprises and industry verticals.
The addition of this edge computing option adds several interesting possibilities of composing and delivering innovative applications and solutions. Many existing distributed applications can get a large capacity and performance boost. For example, the whole multi-player online gaming experience will be enhanced when a part of the logic is processed at a common edge closer to the players.
The below diagram shows the various options of deploying application components and distributing information processing along the path of the data.
5G will provide the technology boost to support URLLC (Ultra Reliable Low Latency Communication) capability to enable use cases such as V2X and Telesurgery, Cobots where the end-to-end latency is expected to be in milliseconds. 5G will also provide eMBB (Enhanced Mobile Broadband) capability for use cases which involve large data rate such as augmented reality and virtual reality.
There are standardization efforts to establish inter-operability for edge computing applications from organizations such as the European Telecommunications Standards Institute (ETSI) and Open Fog Consortium. A set of published API specifications from ETSI are already available.
The European Telecommunications Standards Institute (ETSI) and Open Fog Consortium are collaborating to build edge technologies reference architecture and API specification for 5G. While defining the APIs and enabling their support in the MEC platform is the first step, ensuring their implementation plays a crucial role. The participating VNFs that are used to compose the network slice as required by a use case need to leverage network functions and expose the necessary information through the APIs. Once different players in the ecosystem come together, MEC brings an exciting set of opportunities to realize use cases across a variety of industries, be it easing traffic congestion, telesurgery or connectivity during emergency services.
Subhas Mondal
Subhas is the Chief Architect of the organization-wide 5G initiative at Wipro and has over 27 years’ engineering experience in telecom research & development. He is a Distinguished Member of Technical Staff (DMTS) at Wipro and a Senior Member of IEEE. He is responsible for architecting the Digital Network Transformation theme on the Wipro HOLMES platform.
He is a passionate technologist and has developed a variety of solutions & frameworks in emerging technology areas such as 5G, SDN/NFV, Li-Fi, Cybersecurity and AI for Network Automation. Subhas has also worked at the Centre for Development of Telematics (C-DOT) as a Research Engineer and developed products for remote switching requirements
He is an engineering graduate in Electronics and Electrical Communications from IIT Kharagpur.
Saji Thoppil
Saji has 25+ years IT industry experience in design, build and operationalization of complex and distributed IT systems. He was recognized as a Wipro Fellow, Distinguished Member of Technical Staff, for his distinguished career and contribution to the industry. During his distinguished career, he created multiple practices and incubated many new IPs at Wipro. Wipro’s FluidState Data Center, designed and developed by Saji, was one of the industry’s first blueprints for converged infrastructure. He also architected and deployed Wipro’s first public cloud, which was one of the first public clouds to go live in the Indian sub-continent. Modular remote-managed Edge data centers, liquid cooled and designed for high density was another recognized product launched by him, nominated for many global awards. His recent focus has been on edge computing with a special emphasis on 5G and vertical applications. He is on the governing board of LF-Edge, Oasis TOSCA.