Your team has recently deployed a new version of a service, and you've noticed an unexpected increase in the number of instances within your Kubernetes cluster. The service is configured to scale when the average CPU utilization exceeds 80%.

Which tool would be most effective for investigating this issue?