We’ve received another excellent entry to our Design a Dashboard competition. John Clark has sent us designs for a system overview and a drilled-down view of a selected SQL Server instance.
The overview screen sorts alerts by their status rather than severity, using a ‘traffic light system’ of red,amber,green to identify servers as ‘not acknowledged’, ‘acknowledged’ or ‘ignored’. At this level, there are no details of the types of alert, but only an indication of the number of alerts in each state per server. As John comments:
“I am trying to use the physical location of the icon as the means of prioritizing”
Drilling down into the SQL1 server, the same sorting is applied, but we now see the alert type, the value that triggered the alert, and the alert’s defined threshold value. This kind of sorting encourages the user to at least acknowledge or choose to ignore an alert; an alert is flagged as high priority (red) when it has not yet been acknowledged – “These are things we do not know we don’t know!”
Thanks for submitting these designs John. It’s a very different and thought-provoking approach to how best to prioritize your issues as a DBA. We’ll enter you into the prize draw which will take place very soon…
Would this design suit your workflow? Do you have a novel way of prioritizing your alerts? We look forward to hearing your ideas too.