We are relying on the NLB & Target group metrics and when comparing the data we have in Prometheus using this exporter vs. AWS, it is unclear what time frame the cloudwatch exporter is aggregating the data.
For example, on AWS, we can adjust the aggregation timeframe. But in the cloudwatch exporter, there is no way to do that so it must be preconfigured to aggregate.
Does this depend on the scrape interval?