Ensuring smooth operations of data centres is key to a company’s success. This endeavour is becoming more and more difficult. Given the continuously increasing amount of data, realtime demands and continuous system availability requirements, the Information Technology (IT) infrastructure is increasingly becoming more and more complex and unmanageable. Humans are not able to manually identify in time the breakdowns of the IT systems, let alone predict or avoid them. Hence, there is a need for an overall strategy regarding the early recognition or rather, the avoidance of outages. In the focus of our attention are technologies in the area of artificial intelligence, data analytics, anomaly detection, logging, and parsing, etc. We define an overall strategy in this regard, such that by automating and/or reducing routine activities the support team can concentrate its activity on setting up leading edge technologies rather than relying on the individual skills of some team members in fixing or preventing the failures. To conclude, our strategy supports a paradigm shift from more or less subjectively designed individualistic conceptions in handling of disruptions regarding the service of computer centres and IT infrastructures towards objectively established optimal solutions.