ANTS Profiler^™

Stop walk-through

Step 7: Using the call tree to drill down to the most expensive path

• Steps:
• 1
• 2
• 3
• 4
• 5
• 6
• 7
• 8
• 9
• 10
• Next

The call tree auto-expands to show the worst performing code path within the region we selected on the timeline, taking us, in seconds, straight to the heart of a performance issue. The slowest methods are highlighted by performance bars for quick visual inspection. You can view the profiling results in a number of ways:

• View time in percentage, ticks or milliseconds
• Change between CPU time and wall-clock time, with one easy toggle
• View the call tree top-down or bottom-up
• Hide insignificant methods (those that contribute less than 1% of the total execution time)

Although this is not to be taken too literally, ANTS Profiler suggests candidates for optimization by placing a small red asterisk behind the method it thinks needs to be looked at. These indications tend to be a good starting point if you are not sure where to begin your investigation.

Looking at the results on the call tree, we can clearly see that the method Mandelbrot.Algorithm.EvaluateUsingComplexNumbers only gets hit 205,000 times but takes 77% of the execution time. We can also see that there is a huge drop in the time with children between this method and the one below it, so it seems like we should start by looking into that.

Note: The call tree is dynamically updated to match the selection on the timeline.

BackNext