Performance Issues: Monitor and Analyze CPU utilization on the LAE server

This graph should be used if you're experiencing poor performance on an LAE server.  In order to run it, there are node statistics that need to be turned on that will log the amount of time every node takes to run - see the attached Word document for instructions.

It's purpose is to tell us which nodes are taking a long time to run, those are the ones that are using the most CPU power. 

The results of the graph are intended to be used as a starting point in helping you pinpoint which graphs are using the most CPU power. You can then focus your efforts on making the graph(s) run more efficiently by cleaning up unnecessary items or logically dividing tasks across the server.

FAQ's:

* Are there any known issue due to this change on functionality, performance , etc on the lavastorm jobs? 

functionality - there shouldn't be any issues
performance - there may be some but should be only extremely marginal and not noticeable.

* What would be the rollback option? (Can we remove the added content to config file and reboot the server will revert back the changes?) 

Yes, you can get rid of them or comment them out and restart the server to rollback without issue.

Graph Analysis:

* Depth = Number of nodes running in a single period.  A node is a single process.  So, you could say it is the number of processes running in a single period.

* In the node "Set Period to 5 minutes" qq is an intermediate calculation. This line should be commented out:
`emit qq=floor(tmm/5)`

* The nodes named "Last Month" should be modified to reduce the time period in the chart.
 That node should instead reduce the records to only the period of poor performance

* The node named "Date Filter" also reduces the data to a specific time period.  It reduces it to only one week.

Things to consider during analysis:

* Is LAE the only thing running on the server?

* Do you need to increase CPU's / Memory?