AIMS Anomalies empower users to prevent business disruption  through identify issues early - across thousands to millions of IT components.

 

deep-icon

 

AIMS Anomalies leverage AI and machine learning to provide early notification of IT issues that can bring down your business

AIMS Anomalies allow you to get a comprehensive business wide automated monitoring without the effort required by traditional monitoring solutions.  Anomaly alerts use real time monitoring of actual behavior vs the expected behavior and correlates impact of deviations across your thousands of infrastructure components, applications and services. 

 

Eliminate monitoring blind spots and cry-wolf alerts leveraging effective AI and Machine Learning.

 

 

Learn how it works below

Normal behavior patterns and auto-detection of deviations

Normal behavior patterns are built automatically and continuously updated for all metrics collected by AIMS.    The purpose with the patterns is to dynamically set alerting thresholds based on the actual cyclical behavior of your business - represented by thousands to millions of metrics.  With these patterns you get a high resolution digital DNA of your business.   Deviations are detected in real time as a basis for anomaly alerting (see below).

Custom anomaly groups allow you to instruct AIMS based on your knowledge 

AIMS will be default look for correlation of deviations (as illustrated above) across all "nodes" belonging to a system.  A system is an agent installed to a host or a resource group in Azure.  Using the topology view (shown in the illustration to the right) users can create and edit anomaly groups to instruct AIMS to include groups of systems for Anomalies.  The result is that AIMS will correlate deviations across all metrics for all systems in a correlation group for Anomaly alerts.

Anomalies show performance and business impact

Anomalies are triggered based on a set of criteria including: relative deviation vs expected behavior, climb rate of behavior / how fast the situation escalates and the number of metrics that show correlated deviations.   The Anomaly shows (illustrated to the left) each metric deviation, the system impacted, the time it started, the relative deviation of the deviation vs expected behavior.  This information allows you to easily understand the cause and impact of an Anomaly.

Anomaly Timeline shows the sequence of events

The Anomaly Timeline view is an alternative view of Anomalies that presents Anomalies in the sequence deviations for individual metrics were identified.  This view allows an easier identification of the originating deviation and the follow-on consequences and escalation.  

Drill down to and view further details in Analytics

Investigating an Anomaly further is available from "See in Analytics" shown in the Anomaly and Timeline illustrations above (top right).  Clicking the "See in Analytics" button brings you to a dialogue where you select the metrics you want to dive deeper into.  The data selected will populate a chart in Analytics with the applicable time frame of the anomaly.  This allows you to see how the different metrics escalate and correlate in behavior. 

 

 

AIMS Anomalies automates monitoring through Artificial Intelligence and Machine Learning and replaces static threshold based monitoring.

 

Sign up for the free Community Edition below to test the Anomalies with your own data.