In a Kubernetes cluster, you are tasked with monitoring the performance of your applications and the health of the cluster itself. You decide to implement a monitoring solution that aggregates metrics from various sources, including the Kubernetes API, application logs, and system performance data. After deploying your monitoring stack, you notice that the CPU usage of your application pods is consistently high, leading to performance degradation. To address this, you want to analyze the CPU metrics over a specific time period to identify trends and spikes. If the average CPU usage over a 10-minute interval is recorded as 75% and the maximum usage during that period peaks at 90%, what would be the appropriate action to take based on these metrics?