Performance baselines and trend analysis with PDC
Reactive application health management, where you address issues only after users report them, degrades the user experience. A proactive monitoring strategy identifies and resolves potential issues before they affect users. Use Pega Predictive Diagnostic Cloud (PDC) to monitor Pega applications and maintain performance.
Lead System Architects (LSAs) manage application health and performance. Apply the following strategies with PDC to monitor applications, maintain stability, and optimize the user experience.
Establish a performance baseline
Create a baseline to monitor performance proactively. A baseline is a set of metrics that represent normal application performance. Without a baseline, you cannot measure the impact of new releases or detect performance degradation.
When to establish a baseline
Establish a baseline after a major release stabilizes. This timing ensures accurate measurement of the application state.
What to measure
Focus on key performance indicators (KPIs) that directly affect the user experience. Common KPIs to monitor in PDC include:
- Interaction time: The time it takes for a user to complete a Screen Flow or an Action.
- Browser and service response time: The performance of the browser and the service response times that your application relies on.
- Error rates: The frequency of both client-side and server-side errors.
How to use baselines
Use the baseline as a benchmark for future releases. Investigate any significant deviation from the established baseline.
Monitor trends instead of spikes
Analyze performance trends over time instead of isolated spikes. A single spike might be an anomaly, but consistent trends often indicate underlying issues.
Use historical data
Review historical data in PDC to monitor key metrics over time. Pay attention to trends after a new deployment or a configuration change.
Monitor trends
Watch for the following patterns that indicate performance issues:
- Gradual degradation: A steady increase in interaction times or error rates can indicate a worsening underlying issue.
- Increased volatility: A sudden rise in metric variability can be a sign of instability.
Correlate trends with changes
Link performance trends to application or environment changes to identify root causes.
Prioritize issues with the improvement plan
The PDC improvement plan analyzes system data to identify and prioritize issues that affect performance and stability.
Start with the improvement plan
Create a data-driven, prioritized list of issues for resolution.
Focus on high-impact issues
Address Rules or components that cause the most significant performance issues.
Foster continuous improvement
Review and resolve issues from the improvement plan regularly to integrate performance management into the development lifecycle.
By implementing this proactive monitoring strategy, you shift from reactive to proactive application health management. This approach supports a consistent user experience and promotes a culture of quality and performance.
Check your knowledge with the following interaction:
This Topic is available in the following Module:
Want to help us improve this content?