Intel® VTune™ Amplifier

Viewing a Summary Report

Similar to the Summary window, available in GUI, the Summary report provides overall performance data of your target. The Summary report is displayed by default after running the collect or collect-with actions.

Examples

Example 1: Advanced Hotspots Summary

Display the Summary report for the Advanced Hotspots analysis result r010ah. For hardware event-based sampling analysis results, the summary report includes Results and Platform information, summary per the basic metrics, Elapsed time and CPU Usage, and an event summary.

$ amplxe-cl -R summary -r r001ah

Result Info
-----------
Parameter                 r010ah
------------------------  ---------------------------------------------------------------------
Application Command Line  /home/samples/find_hotspots 
CPU Name                  Intel® Xeon® E5 processor
Computer Name             My Computer
Environment Variables     
Frequency                 2700000000
Logical CPU Count         32
MPI Process Rank          
Operating System          Ubuntu 11.04 
Result Size               7879077
User Name        									My Name

Summary
-------
Elapsed Time:  1.908
CPU Usage:     0.719

Event summary
-------------
Hardware Event Type       Hardware Event Count:Self  Hardware Event Sample Count:Self  Events Per Sample
------------------------  -------------------------  --------------------------------  -----------------
CPU_CLK_UNHALTED.THREAD   4258000000                 2129                              2000000
CPU_CLK_UNHALTED.REF_TSC  3712000000                 1856                              2000000
INST_RETIRED.ANY          6638000000                 3319                              2000000

Use the Elapsed Time metric as your performance baseline to estimate your optimizations. The CPU Usage metric is the total CPU time divided by the Elapsed time, which demonstrates an average value of CPU utilization. For example, 'CPU Usage: 5.907' means near 6 cores was running - average, for your program overall. Probably you have 8 cores CPU, so there are potential to parallelize code

Example 2: Concurrency Summary

Generate a summary report for a concurrency analysis result r002cc. The summary portion of the report shows that elapsed time for the target is 4.385s and the average concurrency is 1.015, which means that a single thread was running during most of the elapsed time.

$ amplxe-cl -R summary -r r002cc

Summary
-------
Average Concurrency:  1.015 
Elapsed time: 4.385

Example 3: Locks and Waits Summary

Generate a summary report for the Locks and Waits analysis result r003lw. The summary portion of the report shows that the multithreaded target spent almost 11 seconds waiting, with an average concurrency of only 1.016. To identify the cause of the wait, view the result in the GUI performance pane, or generate a performance report.

$ amplxe-cl -R summary -r r003lw

Summary
-------
Average Concurrency: 1.016 
Elapsed time: 4.448
Wait time: 10.96

Example 4: Comparison Summary

Comparing two results is a quick way to check for regressions. In this example, a summary report is generated from two consecutive advanced-hotspot results.

$ amplxe-cl -R summary -r r010ah -r r009ah

Result Info
-----------
Parameter                 r010lh                                                                 r009lh
------------------------  ---------------------------------------------------------------------  --------------------------------------------------------------------
Application Command Line  /home/samples/find_hotspots   /home/samples/find_hotspots  
CPU Name                  Intel® Xeon® E5 processor                                          Intel® Xeon® E5 processor
Computer Name             My Computer                                                          		My Computer
Environment Variables                                                                            
Frequency                 2700000000                                                             2700000000
Logical CPU Count         32                                                                     32
MPI Process Rank                                                                                 
Operating System          Ubuntu 11.04                                                    Ubuntu 11.04
Result Size               7879077                                                                8983332
User Name                                                                                        

Summary
-------
               Difference  r010ah  r009ah
-------------  ----------  ------  ------
Elapsed Time:  -0.806      1.908   2.714
CPU Usage:     -0.083      0.719   0.802

Event summary
-------------
Hardware Event Type       Difference:Hardware Event Count:Self  Result 1:Hardware Event Count:Self  Result 2:Hardware Event Count:Self  Difference:Hardware Event Sample Count:Self  Result 1:Hardware Event Sample Count:Self  Result 2:Hardware Event Sample Count:Self  Events Per Sample
------------------------  ------------------------------------  ----------------------------------  ----------------------------------  -------------------------------------------  -----------------------------------------  -----------------------------------------  -----------------
CPU_CLK_UNHALTED.REF_TSC  -2138000000                           3712000000                          5850000000                          -1069                                        1856                                       2925                                       2000000
CPU_CLK_UNHALTED.THREAD   -2732000000                           4258000000                          6990000000                          -1366                                        2129                                       3495                                       2000000
INST_RETIRED.ANY          -3884000000                           6638000000                          10522000000                         -1942                                        3319                                       5261                                       2000000

Example 5: General Exploration Metrics in Summary

The General Exploration analysis metrics provide a high-level overview of hardware performance problems. The General Exploration Metrics section appears in a Summary report for hardware event-based sampling analysis results. The set of metrics displayed in the summary depends on the profiled CPU type and list of events.

Information is displayed by execution categories to make it easy to identify which part of the pipeline to focus on. In this example, the CPI Rate is well over 1, indicating there is a problem. In the Back-end Issues category, LLC Miss (Last-Level Cache Miss) indicates CPU cycles are being spent waiting for LLC load misses to be serviced. View the Hotspots

$ amplxe-cl -R summary -r r001ge

General Exploration Metrics
---------------------------
Parameter                            r001ge     
-----------------------------------  -----------
Hardware Event Count                            
 CPU_CLK_UNHALTED.CORE               16219170376
 INST_RETIRED.ANY                    710657200  
CPI Rate                             22.823     
Retiring uOps                        0.024      
 MS Assists                          0.006      
 FP Assists                          0.0        
 SIMD Assists                        0.0        
Wasted Work                          0.0        
 Branch Mispredict                   0.030      
 SMC Machine Clear                   0.0        
Back-end Issues                      0.0        
 Memory Latency                      0.0        
  LLC Miss                           0.124      
  LLC Hit                            0.0        
  DTLB Overhead                      0.0        
  Contested Accesses                 0.0        
  Page Walk                          0.022      
  Bus Lock                           0.0        
 Memory Reissues                     0.0        
  Split Loads                        0.0        
  Split Stores                       0.0        
  Loads Blocked by Store Forwarding  0.0        
 DIV Active                          0.0        
Front-end Issues                     0.0        
 ICache Misses                       0.0        
 ITLB Overhead                       0.007      

See Also