This triggers an alert upon a sustained utilization greater than 95% of DTU, CPU, Memory, Data IO, or Log.
If this happens for too long, it can cause outages of your Azure SQL DB due to "throttling".
If SQL queries cannot be improved well enough to mitigate this situation, consider scaling up your Azure SQL DB size.
In case of High CPU %:
Use Top SQL and/or QuickTrace to determine which sessions and queries are consuming the most CPU.
NOTE: If you launch QuickTrace by highlighting a range on the CPU Usage chart, the results will be pre-sorted by CPU % descending, so the greatest consumers will be on top.
In case of High Data IO % or High Memory %:
Use Top SQL and/or QuickTrace to determine which sessions and queries are consuming the most Logical Reads or Writes.
In case of High Log IO %:
Use Top SQL and/or QuickTrace to determine which sessions and queries are consuming the most Logical Reads or Writes.
Also check the Query Stats and Procedure Stats for the queries consuming the most Logical Reads, or having the highest Execution Count.
In case of High DTU %:
DTU is an aggregated calculation affected by all resource utilizations (CPU, IO, and Memory).
See the above remediation steps for all resource utilization types.