Logs capture detailed information about an application’s behavior, events, errors. They aid in troubleshooting.
Avoid over-logging (can increase cost for log storage when high volume) and have logging levels. This allows you to
only log densely when a particular flag e.g. DEBUG=true is set.
Metrics are used to understand the health and performance of a system. They are used to aggregate data such as latency, error rates and throughput.