Abstract: Embodiments are directed towards a collection computer that automatically detects and monitors each of a plurality of sensors that are currently providing real-time data regarding characteristics of a first machine. A pattern in the provided real-time data may be determined based on a comparison to another pattern from previously provided real-time data from other sensors associated with a second machine. The comparison is employed to identify an event that previously occurred at the second machine, where a positive comparison may be employed as a prediction that the event corresponding to the second machine is about to happen to the first machine. Based on this prediction, an alert may be provided to at least one user of the first machine. The alert may include information, such as a component has failed, is currently failing, or is about to fail.

Abstract: Embodiments are directed towards enabling users to customize data collection and analysis. A collection computer may automatically detect and dynamically update each of a plurality of sensors that may be currently providing real-time data regarding at least one characteristic of a machine. At least one sensor may be selected for local storage of its corresponding real-time data at the collection computer. And at least one sensor may be selected for remote storage of its corresponding real-time data by a server computer. A template may be employed to remotely display at least one characteristic of the machine based on current real-time data provided by at least one of the sensors identified by the template. In response to the user selecting at least one sensor for remote display, the template may be modified to include the remote display of the at least one sensor's corresponding current real-time data.