Avner
To be able to take a better decision about buying another container/s, or moving to an all different billing method, admins needs to have a clear, global view on the system in terms of "waiting time".
I think first critical tier would be:
  1. Aggregated time of Containers Queue for a month/week, whatever.
Next tier would be:
  1. Make sure #1 above is broken down between nightly and ongoing (e.g. actual dev/deploy waiting for a job to complete)
  2. Average build time, hotspots (specific projects with long running jobs) and critical path analysis.
  3. Anomaly detection (builds that normally took X jump to X *1.5 in the past month).
CCI-I-811