PagerDuty, Inc.
Operations Health Management

Abstract:

Implementations of this disclosure are directed towards the computer-implemented determination of interrupt events from a plurality of notification events generated by one or more of a plurality of computer systems. Interrupt events are associated with notifications designed to immediately alert a responder. A health score is generated as a weighted sum of one or more health sub-scores that are generated using one or more health sub-score models that take as input more health metrics representative of data associated with one or more interrupt events. The health sub-scores are representative of an impact relating to the management of operations of the plurality of computer systems and the weights used in the weighted score are representative of an estimation of the relative impact of the generated health sub-scores. The health sub-score models may be automatically modified according to an automated analysis of historic operational outcome metrics and historic scores.