Published: 07 October 2015
Summary
There are five distinct types of causal analysis that make availability and performance data actionable. While each brings a valid perspective, change-based causal analysis — particularly when combined with Bayesian causal network analysis — holds the most promise for IT operations leaders.
Included in Full Research
- Add Causal Analysis to Pattern Discovery and Anomaly Detection to Make the Data Stored in ITOA Platforms Truly Actionable
- Approach No. 1: Establish a Relationship of Topologically Grounded Causality Among the Variables Describing the System
- Approach No. 2: Situate the Variables Describing the System in a Bayesian Causal Network
- Approach No. 3: Establish the Relationship of Granger Causality Among Variables Describing the System Arranged in a Time Series
- Approach No. 4: Use Pre-Existing, Distributed and Mostly Partial Knowledge of System Causality to Prune the Correlational Variable Networks
- Approach No. 5: Establish What Changes Have Recently Been Introduced to the Environment and Try to Correlate Variable Value Fluctuations to Those Changes
- Combine Bayesian Causal Network Analysis With Change-Focused Analysis for the Most Effective Combination of Causal Analysis Methods