Memory utilization discrepancy between "Network Device Health" dashboard and drill-down Health view for a device

Products

• CA Performance Management - Usage and Administration

Issue/Introduction

We may notice discrepancy in memory utilization values for devices when looking to "Network Device Health" dashboard compared to when we drill-down to specific devices.

For example:

compared to:

Environment

Release:
Component: IMADMN

Resolution

This is expected behavior when comparing a View that shows the Top values per each component (Network Device Health) against a View that shows the average values for all components) on the device. 

The "Top Memory Utilization" View shows the top values for each 'component'. While the "Switch Pages" View when drilling down to the specific device shows the average of 'all' components on that device. 

For example, if there are multiple components for Memory metric on the device, one of them might show 97% for memory utilization, but the value for other component(s) could be lower. This cause the device's Average Memory usage value to be low. 

We could do a Trend View for one specific Device by 'component' for memory utilization, see how many memory components it has, and determine if the values are correct. 
With a View grouping by 'component', we can see if there are multiple components for the same Device name - e.g. for RAM and RAM (Cache); here we should see a line for e.g. 97% and another line (or more) with lower values around 10-20%. This would explain the average value seen for the specific device.