Two Dimensional Time-Series for Anomaly Detection and Regulation in Adaptive Systems

摘要: A two dimensional time approach is introduced in order to classify a periodic, adaptive threshold for service level anomaly detection. An iterative algorithm applied history analysis on this periodic provide the smooth roll-off significance of data with time. The described leads an approximately ten-fold compression storage, and thousand fold improvement computation cycles, compared naive time-series approach. behaviour detector discussed, result implemented cfengine direct use system management.