Ryan Perera, Vice-President and Country Manager, Ciena India

The rising demand for ultra-high-speed networks, coupled with several new-age use cases such as telehealth, immersive learning and online gaming are propelling telcos to rework their urban networks. Further, as telcos stand at the cusp of 5G adoption, it is time for them to own the edge and reimagine metro and edge network architecture. In an interview with tele.net, Ryan Perera, vice-president and country manager, Ciena India, talks about the transformation towards edge networks, how service providers are enabling this transformation, and the way forward…

What factors are leading to the shift towards edge networking? 

The cloud has become an all-pervasive means to deli­ver both business and perso­nal communication services to users. How­ever, most ap­p­lications have been dependent on large, centralised data centre and cloud architectures to deliver such services.  As our dependency on such cloud-based ser­vices grows, along with the need for ne­ar-real-time, low-latency connectivity, the traditional hub-and-spoke model fails to deliver. Applications are increasingly hi­gh­ly latency intensive, making it challenging for a distant central data centre to deliver on expectations. This calls for physical data centres to move closer to where content is both created and consumed.

How will this shift transform the network architecture in urban areas? 

With the exponential rise in data genera­ted by smartphones, cameras, internet of things (IoT) and other connected devices, centralised data centres are under tremendous strain. To deliver the quality of experience that end-users, humans and machi­nes expect today, data centres must be phy­sically located closer to them. This be­comes challenging in an urban set-up whe­re prohibitive real estate costs make it difficult to set up large-scale data centres. Edge data centres are miniaturised versions of centralised data centres and require far less eq­uip­ment, power, and space requirements, making them easier to deploy at the network edge, closer to end-users. However, smaller edge data centres are not in­tended to replace large centralised data centres. Ra­ther, they will complement each other with some data being processed at the edge and some at the core.

What changes do service provi­ders need to make their networks edge-ready? 

Communication service provi­ders (CSPs) must focus on building a meaningful business and technical edge strategy to get the network edge-ready. Rather than just thinking about the right equipment to buy, the edge strategy should consider why you actually need edge computing. Being able to clearly define the technical and business problems that CSPs are trying to solve is an essential part of this exercise – are network constraints the biggest challenge, is it data sovereignty, or is it about reducing your centralised data centre footprint?

Once these and other questions are answered, CSPs can move towards implementation and then start investigating and evaluating available hardware, software and service options. The criteria to consider include cost, performance, features, secu­rity, interoperability and support.

It is important to keep in mind the need for comprehensive monitoring and analytics-based automation. Sending engineers to physically inspect edge data centres is not fiscally prudent or sustainable. Therefore, it is crucial to ensure that the architecture pro­vides for resilience, fault tolerance and self-healing capabilities.

Software tools deployed for edge cloud must include monitoring capabilities, providing a comprehensive over­view of remote deployments; en­able easy provisioning and co­nfiguration; offer compre­h­en­sive alerting and reporting; and maintain the security of the installation and its data.

What new services and use cases will service providers be able to offer to their customers with edge networking? Does it allow them to add new revenue streams? 

The past couple of years have witnessed the emergence of several new applications that are bandwidth and low-latency dependent, including streaming video or content delivery, cashierless retail stores, autonomous driving, industrial IoT and cloud gaming.

Each of these applications requires near real-time latency. Being able to deliver such a high level of service also presents CSPs with new revenue opportunities. Th­is new generation of applications also fa­vours the emergence of new business mo­dels where ev­eryone has a role to pl­ay – mobile and wi­re­line network operators, cloud provi­de­rs, ne­utral hosts, data centre providers, system in­tegrators, and vendors.

What must service providers do to maximise the benefits from this shift?

Edge computing presents CSPs with immense opportunity not just to improve existing business models but to also enter the enterprise space that has been traditionally dominated by IT systems integra­to­rs. For instance, CSPs can bundle IoT services with other enterprise offerings such as private networks. Enterprises do not see CSPs as natural providers of enterprise-centric solutions such as factory automation. Therefore, to become part of an enterprise’s consideration, CSPs must adopt go-to-market strategies including selling edge solutions targeted to meet the specialised requirements of various industry verticals.