Hello @diasgiriaig !
I have seen your question (plus the git issue and email lol 😉
I get your point, I would say that basically the interest here is that you have Splunk here so you can easily take the existing dashboards, modify and/or combine parts of them to achieve the exact result that matches your goal and whishes.
To reply:
- The TOP collection captures various metrics about processes, from CPU to different aspects of memory, IO wait time, etc
- Have a look at the HOWTO interfaces to learn how about how to deal with these metrics with and build sample searches
- Have a look at the TOP processes main user interface and the different dashboard that deal with these (like the bullet dashboard)
You can easily (and the interfaces provided allows you to do) filter on any java processes (process name filter), and perform various kind of calculation or aggregation if you need to so.
Bare in mind that what the TOP process collection provides is what the system sees, not necessary process internal metrics that need to be accessed differently (thinking about JMX in the case of java processes)
But I'm taking note of this, I tried different approaches for this but it is hard from a provider point to be to remain generic enough to be able to answers all use cases while providing best value.
Hope this makes sense !
Guilhem
