Intel® VTune™ Amplifier
Consider performance reports as your starting point for performance analysis. Use them to identify program units (for example: functions, modules, or objects) that take the most processor time (Hotspots analysis), underutilize available CPUs (Concurrency analysis), have long waits (Locks and Waits analysis), and so on.
The reports display the hottest program units in descending order by default, starting from the most performance-critical unit. The command-line reports provide the same data that is displayed in the default GUI analysis viewpoints.
This example opens the Hotspots report for the r001hs Hotspots analysis result. The report shows that matrix is the only candidate for further research and optimization.
$ amplxe-cl -report hotspots -r r001hs -format text -group-by module
Module CPU Time ---------- -------- matrix 8.107
The Hotspots report for the r001hs Hotspots analysis result shows that the algorithm_2 function of matrix took more of the processor time (5.881s). It is the best candidate for further research and optimization.
$ amplxe-cl -report hotspots -r r001hs -format text -group-by function
Function Module CPU Time ------------ ---------- -------- algorithm_2 matrix 5.881 algorithm_1 matrix 2.192 main matrix 0.025 checkResults matrix 0.009
This example displays the hotspots report for the r001hs analysis result including only the top 2 hottest functions in target CPU time. Functions having insignificant impact on performance are excluded from output.
$ amplxe-cl -report hotspots -r r001hs -limit 2
Function Module CPU Time ----------- ---------- -------- algorithm_2 matrix.exe 5.881 algorithm_1 matrix.exe 2.192
Example of a frequency analysis report.
$ amplxe-cl -report frequency-analysis
Frequency Summary for all CPUs
Frequency Turbo Frequency Time (sec)(%) --------- --------------- --------------- 1.596GHz Normal 310.23 (69.15%) 2.661GHz Turbo 138.24 (30.81%) 1.995GHz Normal 0.14 (0.03%) 1.729GHz Normal 0.01 (0.00%) 2.66GHz Turbo 0.00 (0.00%)
Per-CPU
CPU 1.596GHz:Time(sec)(%) 1.729GHz:Time(sec)(%) 1.995GHz:Time(sec)(%) 2.661GHz(Turbo):Time(sec)(%) 2.66GHz(Turbo):Time(sec)(%) ----- --------------------- --------------------- --------------------- ---------------------------- ---------------------------- cpu_0 7.098 (12.66%) 0.012 (0.02%) 0.139 (0.25%) 48.828 (87.07%) 0.000 (0.00%) cpu_1 22.242 (39.66%) 0.000 (0.00%) 0.000 (0.00%) 33.836 (60.34%) 0.000 (0.00%) cpu_2 56.078 (100.00%) 0.000 (0.00%) 0.000 (0.00%) 0.000 (0.00%) 0.000 (0.00%) cpu_3 56.075 (100.00%) 0.000 (0.00%) 0.000 (0.00%) 0.000 (0.00%) 0.003 (0.00%) cpu_4 56.078 (100.00%) 0.000 (0.00%) 0.000 (0.00%) 0.000 (0.00%) 0.000 (0.00%) cpu_5 56.078 (100.00%) 0.000 (0.00%) 0.000 (0.00%) 0.000 (0.00%) 0.000 (0.00%) cpu_6 0.541 (0.97%) 0.000 (0.00%) 0.000 (0.00%) 55.537 (99.03%) 0.000 (0.00%) cpu_7 56.039 (99.93%) 0.000 (0.00%) 0.000 (0.00%) 0.040 (0.07%) 0.000 (0.00%)