In the space of a few short weeks, the COVID-19 outbreak has reshaped the global business environment more effectively than any business calling themselves a “disruptor” could have ever imagined possible.
Businesses of all types had to either institute plans for remote working and access, or rapidly scale them up – reconfiguring business models essential to maintaining business continuity and stability. A challenge even with a steady, years-long roadmap for change. As transformation is catalysed, new technologies that may have been in the pipeline for years are now being looked at as an immediate priority.
For many enterprises, and their supporting communications service providers (CSP), 5G is one of those key technologies. Some may already be advanced in their network rollout and are expanding their use, and some are now accelerating plans to adopt and integrate into their businesses.
For both these types of organisation, the question of network quality definition and maintenance should be front of mind. Departing from how previous generations of connectivity were measured, with 5G, quality will be defined by the customer’s business requirements, not the network’s capabilities (or limitations). In short, it is market driven not technology driven and is about connecting the network to the business. To deliver the service and network quality demanded by enterprise customers, CSPs must first rethink their definition of “quality” — and fundamentally reimagine their 5G network operations.
Guaranteed quality SLAs with network slicing
An end-to-end 5G network delivers on throughput, reliability and latency and different use cases will demand different mixes of these elements. For example, an engineer’s VR headset may require high throughput, very low latency but only a medium level of reliability, whereas a network controlling an autonomous drone may need low throughout, low latency and high reliability. To accommodate this variety a 5G network can be sliced into segments tuned to a specific use case. A network can be sliced into hundreds or even thousands of virtual networks, each with its own custom quality SLA requirements. Enterprises that rely on 5G for their business operations will not tolerate “best effort” approaches and will expect CSPs to deliver guaranteed quality, in other words a ‘committed SLA’. Alongside these multiple use cases which businesses currently rely on, and those yet to come, 5G networks will also incorporate legacy infrastructure in a hybrid network and will, for example, work alongside 4G (LTE networks) which have been around much longer and have wider coverage. Tracking these and ensuring a CSP delivers on multiple SLAs at once is too great a task for operations teams relying on manual processes.
Transform operations to transform service
With one of the main benefits of 5G being its reliability, it’s no surprise to recognise that the service level agreements enterprises expect from CSPs can include guarantees of as much as “six nines” (99.9999%) uptime — and steep penalties for not delivering on those guarantees. That’s a radical departure from the “best effort” approach. However, this is only supportable with CSPs who have highly tuned operations capability.
The multiple use cases, legacy technology and strict SLAs create a new complexity which means quality assurance must go beyond mere network “health checks” and become a fully integrated operational function. This means, effectively, the operations function needs to shift to match new, stricter, customer expectations. Making sure their needs and expectations are met and CSPs can properly control quality across a hybrid network and diverse range of 5G use cases, network and service assurance needs to be holistic and end-to-end, supported by automation and artificial intelligence.
At the foundational level, CSPs should be utilising AI and machine learning to accelerate root cause analysis (RCA) and service impact analysis (SIA). The immediate benefits to operations teams are clear – for example with Nokia’s Assurance Center we have seen a 90% reduction in alarm management using AI-based filtering and correlation, with 30% of issues being resolved automatically without human intervention.
Tapping into an AI ecosystem can also bring the benefit of lessons learned across multiple companies and CSPs. Nokia’s own experience with CSPs using its AVA Cognitive Services bears this out. For instance, these lessons contributed the operations that delivered 75 per cent greater data speeds for Hutchinson 3 in Indonesia.
With the disruption caused by COVID-19, Nokia has been hearing from CSPs and enterprises around the world on the challenges they face. As a result, it has changed its managed service offer for operations to ensure the new pressures on networks, and operations teams, are mitigated. We have seen a demand for these advanced operations services now, in times of stress, and believe the future will further demand for this sort of operations capabilities longer term. CSPs can help keep their networks running by partnering with a managed services provider on business continuity planning (also called contingency planning). With this approach, the alarm monitoring in the CSP’s NOC is shadowed by and mirrored in the managed services provider’s delivery centers. In case of an outage, the managed services provider can use hot-standby switchover processes to step in and take over fault and performance management on a temporary basis — or even take on the full scope of operations, including expanding network capacity to cope with increased traffic.
Shifting patterns of use in both fixed and mobile networks have been stark in recent weeks. The changes have forced a change in transformation plans and timescales. With other elements of business practice and process in flux, it’s vital the communications infrastructure is as dependable as possible. Transforming CSP operations for the world of 5G is the only way to deliver on this new quality expectation.