01-28-2013 08:03 AM
Analytics uses a self-learning algorithm to analyze applications, comparing their behavior to an automatically defined baseline. If Analytics detects behavior which is significantly different from the baseline, Analytics reports the anomaly and identifies the transactions and locations that displayed abnormal behavior. This helps you see at a glance the extent of the business impact of the anomaly.
Analytics identifies probable top causes of the anomaly and similar anomalies, applications that in the past have reported similar issues. This helps you pinpoint the cause of the anomaly and identify a solution.
The Analytics Paradigm
To understand the unique potential of Analytics, you need to change the way that you think of identifying errors:
- Move from a static threshold to dynamic baseline. Instead of a static threshold that needs to be breached to identify a problem, Analytics generates a dynamic baseline which takes into account seasonal information such as busy times of the week or reoccurring patterns. For example if a server generally has lower activity during the night, Analytics may identify abnormal behavior even if the server has low activity but more than is normal for that time of the day.
- Alert only on what matters. Analytics is able to filter out noise, expected behavior, known issues, and identify only the issues that matter to you.
- Isolate the source of the problem. When an anomaly occurs, Analytics can isolate the cause of the problem. If it is a known problem with a known solution, Analytics will suggest a solution. If a similar problem has been identifies as noise in the past, Analytics will be identify this anomaly also as noise.