End-to-End Monitoring at Scale


In order to fulfil service quality and customer satisfaction requirements, Swisscom DevOps teams must make sense of high volume of relevant metrics, events, and traces generated by numerous distributed systems. Behind the scenes, it is necessary for applications to work not only in isolated fashion but also in the end-to-end context, while at the same time quickly addressing all of the issues that could affect the users.

In this talk, we will demonstrate how we empower DevOps teams in Swisscom in their constant effort to ensure stability and continuous improvement of their systems. Our real-time monitoring tool provides root cause analysis and meaningful alarms, which are based on anomaly detection, dependency graphs, and integrated user feedback. In particular, we will show how this was achieved using open source technologies such as Prophet models, PySpark, MLflow, and Prometheus.



Jelena Malic



Jelena is a Data Scientist in the Swisscom’s Data, Analytics and AI department. She recently finished her master studies at EPFL and joined Swisscom firstly as an intern, before becoming a


Joana Soares Machado



Joana is a data scientist at Swisscom. As part of the Data, Analytics and AI department, she develops ML solutions to automate the real-time monitoring of Swisscom’s business processes