Finding relevant dimensions in application service management control

Abstract

The increased interest in autonomous control in Application Service Management environments has driven a demand for analysis of multivariate datasets in this area. This paper proposes a feature selection method using metrics time series analysis. The method exploits four metrics called Similarity, Dependency, Consequence, Interference which are combined in order to perform a multivariate evaluation. This allows more efficient search for similarities in the time-series, selection of most relevant dimensions, and easier control in the reduced space, which would ultimately reduce maintenance effort. This is further used to create causal models of the controlled system, significantly simplifying evaluation of defined elements utilization dependencies. We show that methods based on these metrics can be applied in service control practice under several scenarios. © 2014 Springer International Publishing Switzerland.