When meeting with customers to discuss post-sales support, we often hear the message that maintenance is a legacy service with less added value or importance. In response, I often use the analogy of my expensive road bike. I am an avid cycler, and I can get very excited about a new piece of the latest equipment and the kind of performance boost it can give me. But, in terms of performance, the somewhat less exciting truth is that it is the ongoing maintenance of my equipment that is the most critical to my performance and safety.

Digitalization is upping the ante

Post-launch support, which we call Care, is far from becoming a legacy service. In fact, it is more essential as networks become more complex and the demands and expectations of households and enterprises evolve. Along with increased video streaming and gaming, the pandemic has also seen more people working from home, carrying on video and voice calls and expecting enterprise-levels of reliability and quality of service. Mobile networks are part of the critical national infrastructure, so national regulators often leverage the granting of spectrum licenses to enforce CSP’s service level agreements. And in the age of social media and fast feedback from customers, service degradation and outages can ruin an operator’s brand image and reputation in the time it takes to type 140 characters.

Meanwhile, enterprises (including communications service providers) are ‘digitally transforming’ their operations and embracing Industry 4.0. They are deploying IoT sensors, AI, machine learning and the cloud to create business-critical processes that leverage big data, digital twins and automation. Some will put in private networks, but many others will be interested in buying slices of a public 5G network or deploying hybrid private-public networks. The criticality of these functions will sometimes demand very high performance, ultra-low latency and even 6x9s of reliability, which is a clear focus of maintenance.

Shifting to webscale methodologies

With our customers, we are taking lessons from the IT world by recognizing that open-source software applications can leverage larger community innovations and enable higher agility and speed than proprietary ones. This is a big shift for the telco world, but it comes naturally with the move to software, virtualization and agile development methods. Every customer’s implementation is unique, but they can leverage the efforts of others for more rapid services deployment, fix and upgrade.

From a Care perspective, all of this rapid change means that we need to adapt our maintenance systems for a constantly evolving, customized situation and the exponentially increased complexity of networks. Nokia’s holistic approach to the 5G core is to leverage open-source systems, to create containerized micro-service-based CNFs (cloud-native network functions) that can be managed, integrated and delivered continuously (CI/CD). Nokia’s DelOps (delivery and operations) methodology adapts DevOps to the much more complex operational environment typical of CSPs.

Our solution support addresses all this system complexity and takes the pain of dependency management away from the CSPs. We are also evolving our Care Delivery Model to fully integrate automation and collaboration.

On the automation front, we are coming to rely increasingly on network monitoring to identify potential problems and self-recover before the network service is impacted. We’ve always logged network activity, but did not use the data to its full extent; whereas with machine learning, it is possible to analyze it all. AI allows us to work at the “solution” level, horizontal or vertical, which goes beyond what human perception can achieve in term of data capture and analysis. At one European customer, we used Log Analytics and suspicious patterns to triage a real-time alert and reduce the time to identify the root cause by 60% thanks to our Predictive Network Monitoring (PNM) service reporting a KPI degradation.

Sharing is caring

We create end-to-end teams that use knowledge management systems that enable us to capture specific learnings, quickly identify similar issues using automated systems and share it with our customers — all with the goal of enabling rapid fail, re-integration and re-delivery, continuously. Having to solve network issues should be the exception, but as they say, ‘only unused software has no bugs’. De-bugging is carried out with the full collaboration of integration and delivery.

With 5G and the move to cloud-native, the software-driven virtual approach is changing the way we understand Care. We see our value as being able, on the one hand, to simplify customer operations and increase performance using automation and, on the other, to work collaboratively across our customer base to share our global knowledge for the benefit of all.

We share a common goal with CSPs, which is to provide extraordinary customer experiences under any circumstances. Fulfilling it often feels like shooting at a moving target — given that change is the only constant in this environment. The accelerated pace for rolling out new features and functionalities increases the importance of maintenance to maximize network performance. As an analogy to my cycling trips, my excitement comes from a well-performing bike giving me the boost I was looking for. Nothing is more frustrating during cycling than having to stop to perform repairs that could have been done in advance, well before the exercise.