Hi,
my company produces an enterprise capacity planning suite so I'm very used at seeing charts which overlay performance or workload data with system or application-wide events (here event=incident or human intervention, etc). This way, in front of a strong trend change (cpu or requests per seconds raising up or falling down, ...), I can display my current knowledge at the application level (maintenance on another server in the same cluster, configuration file changes, RAM increase, ...) I can also envision a way to use this to "describe" previous incidents once they've been solved
... View more