Analysis and Research for Network Management Alarms Correlation Based on Sequence Clustering Algorithm

The alarms correlation rules obtained on the bases of network management alarms play an important role on network management and network maintenance. Alarms correlation is a difficult problem in network fault management; sequential pattern mining can be utilized to extract episode rules from network system alarms. This paper introduces the related studies of alarms association and sequential pattern mining; describes the features of network management alarms; presents a method of mining correlation rules of network management alarms by sequence clustering algorithm. The experiments show that many interesting correlation rules could be acquired efficiently. Furthermore, these rules could be used to guide the intelligent network alarms filtering and fault location.