ANOMALY DETECTION AND ALARMING BASED ON CAPACITY AND PLACEMENT PLANNING

摘要

Operational parameters of a running multiserver data processing system are automatically and repeatedly sampled and compared with constraints defined in a capacity and placement planning specification of the multiserver data processing system. Constraint violations are automatically declared as operational anomalies and, if serious enough, corresponding alarm signals are automatically produced.

主权项

1. A machine-implemented anomaly detecting process comprising:
accessing a pre-planned multiserver data processing system comprising a plurality of servers within which there are disposed a plurality of data processing units and a plurality of nonvolatile data storage units; accessing a pre-recorded capacity and placement planning specification for one or more operational layers of the accessed multiserver data processing system, where the operational layers include a physical resources layer comprised of the plurality of servers having disposed therein the data processing units and the nonvolatile data storage units, where the pre-recorded planning specification specifies respective capacity and placement planning constraints respectively for respective components of the one or more operational layers including capacity and placement constraints for components of the physical resources layer; while the multiserver data processing system is running within a pre-planned operational environment and for a given component and a corresponding constraint specified for the component in the accessed capacity and placement planning specification, automatically and repeatedly determining whether the corresponding constraint is violated by a corresponding operational attribute of the given component while the multiserver data processing system is running; and in response to a determined constraint violation, automatically generating and storing a corresponding constraint violation report which identifies the detected constraint violation and identifies a location of the detected constraint violation to a specified degree of location resolution.