What is the slow down anomaly detection timeframe in Dynatrace?

In Dynatrace, the slowdown anomaly detection timeframe is established to identify performance deviations by analyzing historical metrics. Specifically, when the system evaluates the performance of applications and services, it considers a timeframe that is 20% of one week. This timeframe allows the Dynatrace platform to effectively detect anomalies by establishing a baseline for performance during relatively short but representative time durations.

By focusing on 20% of one week, which translates to approximately 1.2 days, the system can capture essential trends and fluctuations without being overly influenced by outliers or irregularities that could come from longer data periods. This specific percentage strikes a balance between having enough data to establish patterns and allowing responsiveness to recent changes in application performance.

Other options reflect different percentages, which would suggest longer or shorter observation periods. However, Dynatrace has determined that a 20% interval is optimal for fast and accurate anomaly detection in terms of slowdown identification. Understanding this concept is crucial for effectively using Dynatrace to monitor and enhance application performance.