Hyperscale and 5G

By 2025, the digital universe is expected to reach 175 zettabytes. The numbers are staggering, but the hyperscale ecosystem is rising to the challenges in meeting the insatiable demand for memory, bandwidth, computing power, storage, and speed. At the same time, cloudification is blurring the proverbial lines between networks and applications, while 5G technologies are pushing more intelligence to the edge of the network. The hyperscale ecosystem and the breakthrough technologies accompanying it are turning these unprecedented challenges into opportunities for operators worldwide.

Hyperscale use cases
The primary 5G use case categories of Ultra Reliable Low Latency Communication (URLLC), Enhanced Mobile Broadband (eMBB), and Massive Machine Type Communication (mMTC) frame a variety of new verticals with unique requirements and performance expectations for latency, densification, bandwidth, speed, and other essential characteristics. Despite the requisite cloudification and redistribution of intelligence from the core to the edge, hyperscale data centers will continue to play a pivotal role in the scalability, SLA conformance, and network efficiency that supports these verticals.

• Factory Automation
  The manufacturing sector has recognized the potential of private edge cloud computing to provide a responsive central nervous system connecting breakthrough robotics, material handling and predictive maintenance applications. This market has expanded as 4G transitions to 5G and private companies leverage URLLC to enable next-generation factory automation or “Industry 4.0”. High bandwidth is required to support smart factory data rates from 1-20 Gb/sec. Densification of real-time IoT sensors can number in the millions for a single deployment. Hand tools as elemental as screwdrivers are now being converted into intelligent devices, continuously streaming torque, position, and calibration data back through the network. Data center software plays a pivotal role in coordinating and analyzing IoT sensor data and optimizing predictive maintenance trigger points.
• Connected Health
  The healthcare industry presents an equally multi-faceted set of verticals encompassing online consultations, remote surgeries, AR/VR training for nurses, telemetry and data management applications, each with disparate latency, bandwidth, privacy and mobility requirements. While routine health check-ups and record transfers may be more forgiving with respect to latency and reliability, cutting edge remote surgical procedures underscore the need for end-to-end network monitoring and assurance. The coordinated data analytics, AI, and machine learning (ML) capabilities of the hyperscale data center and edge computing locations must flawlessly address these flexible requirements to ensure patient safety and mitigate healthcare provider’s liability.
• Unmanned Data Centers
  The next generation of data centers, particularly at the edge, will be overwhelmingly lights out (unmanned). This new reality has accelerated the development of the data center IoT vertical with its own unique and dedicated network slice. The data-center-as-use-case leverages the same real time 5G IoT sensing and remote automation that is redefining manufacturing plants, smart ports, retail centers, and stadium venues. Robots or drones can effectively perform important surveyance tasks within the center, while URLLC can be used to queue automatic links to service dispatch. Much like smart port sensors within shipping containers, strategically deployed temperature and humidity sensors can feed back important environmental data to automate and expedite hardware and HVAC adjustments.

Disaggregated 5G networks
Use cases like these are poised to reshape industries for the better, allowing businesses to dive more granularly into data and chart new paths for efficiency. However, with costs and complexities only set to rise, some of the biggest challenges for hyperscalers will be driven by the intricacies of distributed, disaggregated, cloud-native 5G networks.

With technologies like virtualized RAN, massive multiple input – multiple output (MIMO), and antenna beamforming further complicating radio frequency (RF) and network performance testing, new challenges will likely emerge around spectrum analysis, demodulation, and service-level agreement (SLA) conformance. This end-to-end network complexity is a challenge in a world where operators have to meet higher customer demands for performance, efficiency and reliability, while also managing their own stack.

Network stress testing prevents failures, lowers costs and boosts performance
To overcome this and fully realize the power of 5G and Open RAN, infrastructure must be equipped with seamless end-to-end network slicing orchestrated for the needs of each unique vertical it serves. This requires a departure from legacy methods of data center and network testing and assurance because critical 5G IoT use cases leave no margin for error around reliability and SLA conformance.

New developments in network stress testing can assist in safeguarding or even preventing potential real-world failures. In contrast, pre-deployment testing for 5G verticals could be broadened into RF testing and certification, RAN transport or Xhaul connection calibration and verification, and application, slicing analytics, and emulation. By emphasizing upstream testing and validation, network operators can reduce unplanned outages, troubleshooting or the need for updates after deployment while also boosting performance.