Sameh Yamany, Chief Technology Officer, Viavi Solutions

Hyperscale data centres are continually being challenged by increasing bandwidth, storage, computing pow­er and speed requirements. The rapid scalability that defines hyperscale computing can only be accomplished through a combination of new hardware (horizontal scaling) and the improved performance of ex­is­ting data centres (vertical scaling). Gathering the resour­c­es needed to build or expand hyperscale data centres is a challenge that only grows stronger as the scale increases. This can le­ad to reduced fibre and system verificati­on testing, expo­sing data centres to downstr­eam failures and rework. Given the scale and energy re­­quirements, internet content providers, big data storage and public cloud operators are facing growing pressure to impro­ve efficiency and reduce emissions.

Hyperscale and 5G

5G changes the hyperscale definition

5G has entered the picture with a new blueprint for hyperscale computing. Core functions in the cloud continue to anchor the network architecture, but the need for distributed edge computing and disaggregation to support ultra-low latency 5G use cases pushes hyperscalers out of their pro­verbial box. In other words, 5G big data remains centralised while instant data is moving closer to the edge.

Hyperintelligent hyperscale

Intelligence and automation are needed to successfully create and test 5G network sli­ces from end to end. A successful union between 5G and hyperscale will require artificial intelligence (AI), machine learning, and network function virtualisation in order to achieve performance.

5G hyperscale use cases

Advanced driver-assistance systems (ADAS)

ADAS have established a new transportation model, with 5G providing the requisite ultra-reliable low-latency communication. Edge computing power is the key to meeti­ng ADAS latency requirements. Pa­ra­me­t­ers such as vehicle spacing, traffic sig­nal ti­m­i­ng, pedestrian avoidance and augmented si­g­nage can be fully automated and optimised.

Factory automation

The benefits of factory automation backed by high-bandwidth, low-latency private 5G networks are seemingly limitless. Robots, ve­­hicles, facilities and tools can become sm­a­rter, safer and more efficient, while ma­in­t­e­nance and calibration can be scheduled ba­s­ed on feedback from millions of embedded sensors utilising hyperscale cloud computing.

Connected health

Telemedicine can provide a path to routine care for isolated, immobile or symptomatic patients. The internet of things (IoT) wearable market is set to explode with the capacity boost and latency reduction provided by 5G. Hyperscale data centres that are in perfect sync with edge co­m­puting locations are the key to supporting these virtual healthcare applications securely and reliably.

Unmanned data centres

The use of IoT to monitor and control tem­perature, power and surveillance functions in real time is in line with the shift towards lights-out (unmanned) data centre operations, particularly at the edge. Remo­ving manual operations also opens up new possibilities for hyperscale data centre locations, including frigid, inhospitable regions where land and natural cooling sources are inexpensive and plentiful.

Resolving the hyperscale challenges

Harnessing and embracing new technology (including 5G) to test, monitor and str­eamline data centre operations is the best way to turn the current challenges into op­portunities. Despite the emphasis on 5G radio access networks (RANs) and devi­ce innovations, comprehensive testing of hyperscale data centres is also necessary to ensure the promise of 5G. A proactive app­roa­ch to pre-deployment of fibre RAN and crosshaul testing would utilise a new standard of automated cloud-based testing and diagnostics tools that help rather than hinder construction timelines. Such a progressive approach to testing would also include live network traffic emulation and AI-powered “self-healing” capabilities to prevent out­ages, repairs and unplanned updates.