UsePerfData is enabled by default by hotspot, even though the collected performance data is rarely used.

Maybe we should disable the data collection by default??

0.038613 - 0.036185 = 2.43 ms
1.0 - (0.036185 / 0.038613) = 6.29%

===========

Performance counter stats for 'bin/java -XX:+UsePerfData -version' (100 runs):

             52.95 msec task-clock # 1.371 CPUs utilized ( +- 0.62% )
               205 context-switches # 0.004 M/sec ( +- 0.99% )
                12 cpu-migrations # 0.232 K/sec ( +- 1.91% )
             2,690 page-faults # 0.051 M/sec ( +- 0.05% )
       128,363,276 cycles # 2.424 GHz ( +- 0.16% )
        80,560,141 stalled-cycles-frontend # 62.76% frontend cycles idle ( +- 0.22% )
        63,528,144 stalled-cycles-backend # 49.49% backend cycles idle ( +- 0.26% )
        95,515,717 instructions # 0.74 insn per cycle
                                                  # 0.84 stalled cycles per insn ( +- 0.12% )
        18,532,475 branches # 350.026 M/sec ( +- 0.13% )
           968,290 branch-misses # 5.22% of all branches ( +- 0.15% )

          0.038613 +- 0.000335 seconds time elapsed ( +- 0.87% )


 Performance counter stats for 'bin/java -XX:-UsePerfData -version' (100 runs):

             51.93 msec task-clock # 1.435 CPUs utilized ( +- 0.46% )
               204 context-switches # 0.004 M/sec ( +- 0.96% )
                13 cpu-migrations # 0.247 K/sec ( +- 1.99% )
             2,678 page-faults # 0.052 M/sec ( +- 0.04% )
       126,073,436 cycles # 2.428 GHz ( +- 0.14% )
        79,016,450 stalled-cycles-frontend # 62.67% frontend cycles idle ( +- 0.18% )
        62,262,828 stalled-cycles-backend # 49.39% backend cycles idle ( +- 0.22% )
        93,900,452 instructions # 0.74 insn per cycle
                                                  # 0.84 stalled cycles per insn ( +- 0.13% )
        18,227,636 branches # 350.991 M/sec ( +- 0.13% )
           956,777 branch-misses # 5.25% of all branches ( +- 0.15% )

          0.036185 +- 0.000221 seconds time elapsed ( +- 0.61% )