With performance monitoring, Sentry tracks application performance, measures metrics like throughput and latency, and displays the impact of errors across multiple services. Sentry captures distributed traces consisting of transactions and spans to measure individual services and individual operations within those services.
The Performance page is the main view in sentry.io where you can search or browse for transaction data. The page displays graphs that visualize transactions or trends, as well as a table where you can view relevant transactions and drill down to more information about them.
Using the information on this page, you can trace issues back through services (for instance, frontend to backend) to identify poorly performing code. This allows you to answer questions like whether your application performance is getting better or worse, if your last release is running more slowly than previous ones, or which specific services are slow. Then you can begin to address the code that’s degrading performance.
The Performance page provides you with several filter and display options so that you can focus on the performance data you want. Search conditions help you find the transactions you want to investigate. You can also control what kind of information is visualized in the graphs or which transactions appear first in the transactions table.
The trends view surfaces transactions that have had significant changes in their performance over time. This view is ideal for providing insights when you have transactions with large counts.
Distributed tracing provides a connected view of related errors and transactions by capturing interactions among your software systems. With tracing, Sentry tracks your software performance and displays the impact of errors across multiple systems. Tracing issues back through services connects your front-end to your back-end.
The Trace View allows you to drill down into the details of a single trace and traverse every transaction in that trace. This accelerates your ability to debug slow services, identify related errors, and root out other bottlenecks. You can also drill all the way down into a span for a single transaction and information about it on the Event Detail page.
When you find a transaction of interest, you can investigate further by going to its Transaction Summary page. Every transaction has a summary view that gives you a better understanding of its overall health. With this view, you'll find graphs, instances of these events, stats, facet maps, related errors, and more.
The summary page for Frontend transactions has a "Web Vitals" tab, where you can see a detailed view of the Web Vitals associated with the transaction. You can access a Transaction Summary page from the transactions table on the Performance page.
There are several types of metrics that you can visualize in the graphs, such as Apdex, Transactions Per Minute, P50 Duration, and User Misery to get a full understanding of how your software is performing.