You shall easily check this using PLA(Pegarules log analyzer). This gives you a comprehensive overview of errors/exceptions/alerts and you shall filter by alerts(for example pega 0005 indicates database operation exceeded the threshold, Pega0011 indicates service time..etc)..
If your application is running on cloud, you shall leverage PDC to get a comprehensive view of these alerts/performance issues.
A summary of all the interaction is stored in the pr_perf_stats table (Log-Usage) class. Anything that exceeds the threshold would generate an alert. Alerts and Log-Usage are correlated by AES / PDC to give a more comprehensive picture.