The marina-permit real-time variant is even on CPU, but each host is drowning in invoke count at noon. The scaling signal should be traffic per host, not a guess at CPU. Which auto scaling metric fits?
Select an answer to reveal the explanation.
Short Explanation
The marina-permit variant is even on CPU but drowning in noon invoke count. Scale on InvocationsPerInstance. CPU is the wrong signal when hosts are even, and a Domain 4 latency alarm is not this scaling choice.
Full Explanation
Choose invocations per instance when load is request-count bound. CPU is the wrong signal when hosts are even on CPU and drowning in invokes. Translate is not that metric, and a Domain 4 latency alarm is not this scaling choice.