What′s New in dotTrace

The Monitoring tool window in Rider has been reworked to provide a more visual and integrated profiling experience. It now includes interactive charts for CPU usage, memory consumption, and garbage collection (GC) activity, giving you an at-a-glance view of your application's runtime behavior.

From the tool window, you can initiate performance or memory profiling sessions for deeper analysis in the built-in dotTrace and dotMemory profilers or their standalone counterparts. On Windows, the Monitoring tool also automatically detects performance bottlenecks, UI freezes, and GC issues. Detected problems and time intervals selected on the charts can be further investigated in dotTrace.

Learn more

In this release, we’re introducing tabs to dotTrace. You can now open the same snapshot in multiple tabs with different filter configurations. This lets you examine various aspects of your performance data simultaneously. You can analyze one dataset in one tab while applying different filters in another and switch between views without resetting your filters.

We’ve introduced comprehensive support for the Undo and Redo actions and filter history in the standalone version of dotTrace. Now, each filter change is recorded, making it easy to revert to any previous state. Use the Undo button or hotkey to undo the latest change and the Redo button to reapply it. You can also access a drop-down list to revert multiple steps at once, with changes listed in chronological order.

• dotTrace Viewer now uses a new rendering engine across all platforms. If you encounter issues with the updated viewer, you can disable the Use new cross-platform dotTrace Viewer by default option in dotTrace Home | Settings.
• The standalone version of dotTrace for macOS and Linux gets the Options window.
• We fixed the issue with missing kernel events on Windows: Kernel sampling events were missing in Timeline snapshots due to interference from Windows Defender.
• We've also made other smaller improvements and bug fixes to enhance usability.

Now, you can profile applications based on the MAUI framework.

Currently, dotTrace supports:

• macOS – Mac Catalyst applications targeting .NET 7.0+. Timeline profiling only.
• Windows – WinUI applications targeting .NET 7.0+. All profiling types are supported. Attaching to already running applications is available only for the Timeline and Sampling types.

The support is available for dotTrace standalone and dotTrace integrated into Rider and ReSharper.

dotTrace now includes user interface (UI) presets. The presets adjust the UI to show only the profiling settings relevant to specific applications: General (all application types), Unity developer, and Unreal Engine developer.

We improved profiling on macOS:

• Now, the standalone version of dotTrace lets you specify application bundles (.app) instead of executable files.
• dotTrace in Rider can profile macOS run configurations. The following target frameworks are supported: net7.0-macos, net8.0-macos, net7.0-maccatalyst, net8.0-maccatalyst, Xamarin.Mac.

Now, when configuring a Timeline profiling session, you can specify the following advanced options:

• Sampling rate – Lets you find the optimal balance between the performance of the profiled application and data accuracy.
• Sample allocations every … KB for native allocations – Lets you improve the performance of snapshot analysis by reducing the amount of collected data, i.e. snapshot size.
• Download in background option for native symbol files – Downloads symbol files during the profiling session.

Instead of analyzing the Call Tree, you can now view performance profiling data right in the source code. The Source view displays time distribution information for specific lines of code. As of now, this feature is only available for timeline snapshots taken on Windows.

Now, the Call Tree view lets you group call stacks by individual threads, allowing deeper insight into thread-specific performance issues.

The standalone version of dotTrace is now available on Linux and macOS. Now, you can perform the full profiling workflow on these operating systems, including collecting and analyzing performance snapshots.