Visible to Intel only — GUID: GUID-C728FA97-AC36-43EE-AEF7-16DB8B125611
Visible to Intel only — GUID: GUID-C728FA97-AC36-43EE-AEF7-16DB8B125611
anomaly-detection Command Line Analysis
Use Anomaly Detection to identify performance anomalies in frequently recurring intervals of code like loop iterations. Perform fine-grained analysis at the microsecond level.
This is a PREVIEW FEATURE. A preview feature may or may not appear in a future production release. It is available for your use in the hopes that you will provide feedback on its usefulness and help determine its future. Data collected with a preview feature is not guaranteed to be backward compatible with future releases.
Syntax
vtune -collect anomaly-detection [-knob <knobName=knobValue>] [--] <target>
Knobs:
- ipt-regions-to-load
Specify the maximum number (10-5000) of code regions to load for detailed analysis. To load details efficiently, maintain this number at or below 1000.
Default value : 1000
Range : 10-5000
- max-region-duration
Specify the maximum duration (0.001-1000ms) of analysis per code region.
Default value : 100 ms
Range : 0.001-1000ms
For the most current information on available knobs (configuration options) for Anomaly Detection analysis, enter:
vtune -help collect anomaly-detection
Example
This example shows how to run Anomaly Detection analysis on a sample application called myApplication. The analysis runs over 1000 code regions, analyzing each region for 300 ms.
vtune -collect anomaly-detection -knob ipt-regions-to-load=1000 -knob max-region-duration=300 -- /home/test/myApplication
If you want to transfer the collected data to a different system for analysis, you must archive the result by moving all related binaries to the result folder. After Anomaly Detection analysis completes, run this command:
vtune -archive -r <location_of_result>
What's Next
When the data collection is complete, do one of the following to view the result:
Use the -report action to view the data from command line.
Use the -report-output action to write report to a .txt or .csv file
Open the data collection result (*.vtune) in the VTune Profiler graphical interface.