For infrastructure metrics like CPU, Heap, GC, Hung Thread, we can get the overall host health, but i wish to get JVM-wise utilization of above mentioned metrics in a single dashlet.
Suppose i am seeing high CPU utilization in a graph, i would like to get the JVM or instance name for causing high CPU in that particular dashlet itself. so that i get directly address to that particular instance only. Otherwise i need to check for all jvm to see which one is causing issue.
Could anyone suggest on this please.
Thanks in advance
I recommend charting all JVM measures of interest on a single dashboard so you can view all JVMs in one place and look for outliers among the various measures.
I usually do this with an area chart for each measure, and I also like using a dynamic measure matrix for this type of information with a visualization of "bar" set for each column to allow easy sorting and visualization of the measures.
If you want a few custom dashboards of this type to see if they work for you, just send me an email (dave dot mauney at dynatrace dot com).
Thanks a lot @Dave M. for your quick response. I do have more than 200 JVMs which is very difficult to chart on a single dashboard that is why I was looking for some alternative to get a glance of al JVM in a single dashboard so that I would pay attention only on issue causing instance.
Appreciate your help
How about setting thresholds and using a traffic light for each measure of interest? The dynamic measure matrix has a "Value and Threshold Violation" visualization that would help with showing a lot of measures for a lot of JVMs easily.