I would strongly recommend downloading the Alteryx Server Usage workflow to track scheduler & server activity.
This will give you good visibility to those workflows which are causing the most issues and also taking up most of the resources.
This can help you manage and investigate these jobs, with an aim to optimize these workflows.
The aim here would be then to identify if you potentially need to look at scaling out worker nodes to process the high volumes of scheduled jobs you have running. Again, metrics should as workflow length, workflow make=up (predictive processes which are memory intensive) can all be derived from the above practices.